sign in sign up
<<
Home , Arteriosclerosis, Atheroma, Atherosclerosis, Claudication, Diabetic foot ulcer

CHAPTER 20 PERIPHERAL ARTERIAL DISEASE, FOOT ULCERS, LOWER EXTREMITY , AND Edward J. Boyko, MD, MPH, Matilde Monteiro-Soares, DPM, PhD, and Stephanie G.B. Wheeler, MD, MPH

Dr. Edward J. Boyko is Professor at the University of Washington and Staff Physician at the Veterans Affairs Puget Sound, Seattle, WA. Dr. Matilde Monteiro-Soares is an Investigator at the Center for Health Technology and Services Research (CINTESIS) and an Instructor in the Department of Community Medicine Information and Health Decision Sciences (U753-FCT), Oporto Faculty of Medicine, Oporto, Portugal. Dr. Stephanie G.B. Wheeler is Associate Professor at the University of Washington and Staff Physician, Veterans Affairs Puget Sound, Seattle, WA.

SUMMARY

Peripheral arterial disease (PAD) is common among persons with diabetes, and estimates of prevalence range from 10% to 20%. The condition is often . Persons with diabetes are at increased risk for PAD and often have more distal than persons without diabetes. PAD is associated with substantial morbidity, including pain and functional impairment, , and higher risk of . (DFU) occurs commonly in persons with diabetes, with a lifetime prevalence estimated between 12% and 25%. Healing of DFU may take months to years, and often these lead to lower extremity amputation (LEA). The leading cause of DFU is neuropathy, with contributions from multiple other risk factors, including PAD, diabetes duration and control, and self-care factors. Although diabetes accounted for the majority of all LEA in the United States in 1997, the frequency of hospitalizations for amputation among persons also coded as having diabetes fell dramatically between 1996 and 2008, from approximately 11 to 4 per 1,000 persons hospitalized. Although reduced, this rate is approximately sevenfold higher compared to persons without diabetes. PAD, DFU, and LEA have a considerable negative impact on both the functional status and survival of persons with diabetes in the United States.

PERIPHERAL ARTERIAL DISEASE

INTRODUCTION endothelial cell dysfunction, abnormalities development. Peripheral arterial disease (PAD) refers in vascular smooth function, activates inflammatory mediators and to partial or complete obstruction of the abnormalities, and a hyperco- reactive products that are associ- peripheral , typically the arteries agulable state (1,2,3). (NO) ated with abnormal migration of vascular in the legs. The most common symptom is an important mediator of endothelial cells, so that advanced of PAD is intermittent , which function due to its effects on vasodilation, atherosclerotic lesions in diabetic patients is calf and lower extremity pain that leukocyte-vascular wall interactions, and have fewer cells develops with walking or other exertion platelet aggregation. Hyperglycemia compared with lesions in those without and is relieved by rest. However, the contributes to the loss of NO homeostasis diabetes. This can promote atherosclerotic majority of persons with PAD are asymp- by blocking endothelial cell NO synthase formation and plaque instability. tomatic. PAD is more common among (4) and increasing production of reactive Glucose entry into is not depen- persons with diabetes due to the higher oxygen accompanied by vascular dent on , so glucose levels in the risk for arterial associated (5). leads platelet are similar to intravascular levels with this metabolic disorder. to increased free fatty acid levels, which and can lead to changes associated with activate protein kinase C, inhibit phos- accelerated atherogenesis, including oxida- PATHOPHYSIOLOGY OF phatidylinositol-3 kinase, and increase tive stress, increased platelet aggregation, ATHEROSCLEROSIS IN DIABETES production of reactive oxygen species. and decreased levels of endogenous inhib- Diabetes is associated with an increased Increases in these proinflammatory factors, itors of platelet activity. Hyperglycemia risk for atherogenesis and vascular together with the loss of NO homeostasis increases coagulability and impairs inflammation, caused by hyperglycemia, and increased local , are fibrinolysis through increased production excess free fatty acids, insulin resistance, associated with the transformation of of , a potent procoagulant. and other factors. Inflammation and leukocytes into foam cells (3). Transition Hyperglycemia also increases plasma atherogenic activity are associated with to foam cells is an important early step in concentrations of factor VII and

Received in final form January 6, 2015. 20–1 DIABETES IN AMERICA, 3rd Edition plasminogen activator inhibitor type 1 and TABLE 20.1. Mean Ankle-Brachial Index Among Adults Age ≥40 Years, Overall and by decreases endogenous , Diabetes Status, U.S., 1999–2004 such as antithrombin III and protein C. MEAN (STANDARD ERROR) Diagnosed Undiagnosed Differences in Pathophysiology Overall All Diabetes Diabetes Diabetes No Diabetes Compared to Persons Without Diabetes Right side 1.12 (0.003) 1.10 (0.005) 1.09 (0.006) 1.10 (0.012) 1.13 (0.004) Arterial disease in people with diabetes is Left side 1.13 (0.003) 1.11 (0.005) 1.11 (0.006) 1.11 (0.013) 1.13 (0.003) both morphologically and physiologically Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose different than in persons without diabetes ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. (6,7,8). The femoropopliteal arterial A1c, glycosylated hemoglobin. All relative standard errors ≤30% segments are most often affected, as in SOURCE: National Health and Nutrition Examination Surveys 1999–2004 nondiabetic patients. However, smaller vessels below the knee, including the FIGURE 20.1. Prevalence of Peripheral Arterial Disease Among Adults Age ≥40 Years, by profunda femoris, popliteal, anterior tibial, Diabetes Status, U.S., 1999–2004 peroneal, and posterior tibial arteries, are more severely affected in diabetic than in Overall All diabetes Diagnosed diabetes nondiabetic patients (2,8,9,10), with a high Undiagnosed diabetes No diabetes prevalence of diffuse rather than focal 18 lesions. In addition, medial calcification 15 of the tibial and peroneal arteries is more 12 common. Diabetes is associated with a propensity to earlier arterial calcification, 9 increased thrombogenicity, and generally 6 poorer prognosis. 3

Peripheral arterial disease (%) 0 DEFINITION AND Diabetes Status MEASUREMENT OF PAD Peripheral arterial disease is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on PAD refers to narrowing of the vascular self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions resulting in a reduction in blood for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% supply that leads to inadequate oxygen- confidence intervals. See Table 20.2 for further details. A1c, glycosylated hemoglobin. All relative standard errors ≤30% ation of the tissues of the lower extremity. SOURCE: National Health and Nutrition Examination Surveys 1999–2004 The most common cause of PAD is athero- sclerosis, in which the arterial lumen Noninvasive Measurement cutoff of 0.80. Severe obstruction requiring becomes occluded by plaque arising from The ankle-brachial index (ABI) is a noninva- evaluation is usually the intima. This process largely affects the sive, simple to perform, inexpensive, and recommended for a value <0.4 or <0.5. large and medium-sized arteries, usually widely used method for the assessment of at branch points and bifurcations (10,11). arterial blood flow to the lower extremity. In persons with diabetes, calcification of The ABI is measured in a supine patient by the tibial and peroneal arteries may render Invasive Measurement obtaining the brachial and ankle systolic them noncompressible and produce a Visualization of the arterial vasculature is pressures using a 5–7 MHz handheld falsely elevated ABI considerably greater possible via radiographic contrast angi- Doppler device. Both the posterior tibial than 1.0 (10). Symptoms of PAD may ography, which is considered the gold and dorsalis pedis systolic pressures should therefore occur even with an ABI >0.9, standard for vascular disease diagnosis. be obtained, because adequate flow in if noncompressible, calcified vessels Due to its invasive nature and the risk either of these arterial beds is sufficient to result in falsely high readings of the ankle of kidney injury, radiographic contrast perfuse the foot. The ABI is calculated by systolic (14). The phenom- is rarely used in clinical diag- dividing the higher of the posterior tibial enon of greater frequency of calcified, nosis other than in the setting of planned or dorsalis pedis systolic pressures by the noncompressible arteries in diabetes may revascularization, where it is performed to higher of either of the brachial systolic explain the similar values for mean ABI precisely localize anatomic arterial obstruc- pressures (12,13). The normal ABI can be by diabetes status seen in new analyses tions. Magnetic resonance angiography is defined as 0.9–1.3 1( ). To account for vari- conducted for Diabetes in America, 3rd fast becoming an important noninvasive ability in the measurement, it is generally edition, based on the National Health and method for the detection of PAD. agreed that a lower cutoff value of 0.95 is Nutrition Examination Surveys (NHANES) normal (10). PAD is often defined as an ABI 1999–2004 (Table 20.1), despite a higher ≤0.90, although some studies have used a prevalence of ABI <0.9 among persons with diabetes (Figure 20.1, Table 20.2).

20–2 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

TABLE 20.2. Prevalence of Peripheral Arterial Disease Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Status, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) CHARACTERISTICS Overall All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes Overall 5.23 (0.38) 10.99 (1.40) 10.79 (1.64) 11.39 (2.55) 4.32 (0.43) Age (years) 40–64 2.72 (0.35) 5.68 (1.31) 5.11 (1.14) 6.88 (2.96)2 2.36 (0.39) 65–74 11.27 (1.43) 16.84 (3.32) 17.99 (3.67) 14.47 (6.33)2 9.57 (1.55) ≥75 15.27 (1.46) 22.93 (3.09) 21.88 (4.36) 25.18 (6.57) 13.40 (1.58) Sex Men 4.72 (0.48) 11.40 (1.81) 11.15 (1.97) 11.85 (3.43) 3.47 (0.49) Women 5.70 (0.56) 10.48 (1.73) 10.41 (2.01) 10.67 (3.99)1 5.06 (0.61) Race/ethnicity Non-Hispanic white 5.13 (0.46) 12.13 (1.95) 11.83 (2.33) 12.70 (3.40) 4.18 (0.50) Non-Hispanic black 7.99 (0.95) 14.77 (1.63) 15.68 (2.35) 12.20 (2.98) 6.30 (1.06) All Hispanic 4.13 (1.09) 5.82 (1.72) 5.07 (1.82) 3 3.70 (1.03) Mexican American 3.76 (0.79) 5.81 (1.42) 6.26 (1.13) 3 3.23 (0.94) Smoking status Never smokers 3.41 (0.48) 5.39 (1.09) 5.97 (1.42) 4.10 (1.63)1 3.12 (0.54) Former smokers 6.64 (0.77) 14.98 (2.33) 15.83 (3.19) 13.42 (4.28)1 5.09 (0.87) Current smokers 6.98 (0.93) 15.74 (3.94) 12.20 (2.32) 24.46 (10.32)2 5.76 (0.91) Peripheral arterial disease is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

An American Diabetes Association features of PAD clinical symptoms: pain Measures Suitable for consensus statement recommended using located in one or both calves, provoca- Epidemiologic Research the ABI to screen for peripheral vascular tion by walking quickly or uphill, never The ABI is a valid and reproducible disease in persons with diabetes age >50 occurring at rest, forces the subject to measurement of PAD. Compared with an years (15). The issues of screening and stop or slacken pace, disappears within assessment of or a medical history, misclassification and the limitations of the 10 minutes of rest, and never disappears the ABI is more accurate (1). The ABI has ABI were acknowledged. However, the with continued walking. The original Rose been validated against angiography and problems were not felt to detract from the questionnaire has only moderate sensi- found to be 95% sensitive and almost clinical usefulness of the ABI to screen tivity (60%–68%) in capturing persons 100% specific 19( ,20). Limitations to the for and diagnose PAD in persons with with this clinical diagnosis (18) when ABI are that it is inaccurate in patients diabetes. Hallux pressures may be used in physician diagnosis and ABI are used with calcified, poorly compressible patients with medial calcification in as the gold standard. The Edinburgh vessels and in symptomatic patients with whom the ABI is elevated. Calcification of Claudication Questionnaire, a simplified moderate aortoiliac stenoses (1). the arterial media is common in persons version of the Rose questionnaire, has with diabetes, but medial calcification improved diagnostic test indices with EPIDEMIOLOGY does not extend into the digital arteries. sensitivity of 91.3% (95% confidence Data Sources and Limitations Thus, perfusion pressure can be assessed interval [CI] 88.1%–94.5%) and specificity One general data source for this chapter by measuring hallux systolic pressure of 99.3% (95% CI 98.9%–100%) (18). includes new analyses of existing U.S. using either a strain-gauge sensor or national health survey data conducted photoplethysmography (16). Exercise Testing for Diabetes in America, 3rd edition. Exercise testing can help with diagnosis These surveys include self-reported data Symptom-Based Diagnosis of PAD in patients who have typical symp- from telephone or in-person interview of Claudication is an insensitive measure toms of claudication but a normal ABI or participants and, in some cases, physical of PAD, with symptomless diminished in those with atypical symptoms. Patients examination and laboratory and imaging arterial flow estimated to occur at walk on a graded treadmill until symp- studies. Strengths of such surveys are least two to five times as frequently as toms are elicited, and ABIs are recorded that they have national representation. symptomatic claudication (17). Multiple immediately thereafter (10). Patients with Limitations include the inaccuracies of questionnaire instruments are available arterial obstruction will typically have a self-reported information and reliance on to assess the presence of claudica- >20 mmHg drop in ankle pressure after measurements other than the reference tion, including the Rose questionnaire exercise. standard, due to the cost and sometimes (11), which inquires about the following invasive nature of such tests. Other

20–3 DIABETES IN AMERICA, 3rd Edition

TABLE 20.3. Prevalence of Intermittent Claudication Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Smoking Status, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) CHARACTERISTICS Overall All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes Overall 14.13 (0.68) 23.21 (1.17) 27.72 (1.53) 12.80 (2.55) 12.52 (0.76) Age (years) 40–64 13.59 (0.83) 23.59 (2.19) 29.95 (2.84) 9.81 (2.58) 12.25 (0.89) 65–74 16.34 (1.63) 24.67 (2.48) 25.74 (2.85) 21.95 (5.44) 13.51 (2.01) ≥75 14.61 (1.44) 19.72 (2.77) 23.44 (3.28) 10.40 (3.51)1 13.19 (1.55) Sex Men 12.26 (0.77) 19.27 (1.75) 24.93 (2.14) 8.66 (2.37) 10.84 (0.87) Women 15.76 (1.08) 27.55 (1.78) 30.40 (2.28) 19.07 (4.36) 13.92 (1.22) Race/ethnicity Non-Hispanic white 13.23 (0.86) 21.85 (1.68) 26.84 (2.29) 11.68 (2.93) 11.92 (0.91) Non-Hispanic black 19.64 (1.83) 33.16 (2.67) 36.72 (2.98) 21.61 (6.02) 15.84 (2.18) All Hispanic 16.96 (1.86) 23.07 (2.17) 25.49 (2.14) 15.20 (5.47)1 15.32 (2.36) Mexican American 17.40 (1.29) 26.56 (2.27) 30.48 (2.24) 15.67 (4.62) 14.94 (1.63) Smoking status Never smokers 11.76 (0.80) 21.04 (1.64) 26.59 (2.21) 6.95 (1.93) 10.16 (0.92) Former smokers 13.29 (1.12) 22.59 (1.90) 25.79 (2.53) 16.13 (4.17) 11.43 (1.35) Current smokers 21.04 (2.01) 30.37 (3.63) 34.30 (4.16) 20.21 (6.69)1 19.66 (2.19) Intermittent claudication is defined as answering yes to both of the following questions: “Do you ever get pain in either leg while you are walking?” and “Does this pain include pain in your calf or calves?”. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% SOURCE: National Health and Nutrition Examination Surveys 1999–2004 sources of data include published reports Unadjusted prevalence varied consider- diabetes, possibly due to longer or more of investigations conducted in other popu- ably by diabetes status, with greater than severe disease leading to higher risk of lation or clinical settings. Limitations of a twofold difference seen in persons with atherosclerosis. This result differs when such data include limited generalization, diagnosed and undiagnosed diabetes PAD is defined using ABI <0.9 (Figure potential selection bias, and at times, low compared to those without diabetes 20.1, Table 20.2), where a higher preva- power due to smaller sample size. (Figure 20.1, Table 20.2). In the NHANES lence of PAD is seen in both known and 1999–2000 among adults age ≥40 years undiagnosed diabetes compared to nondi- Prevalence of PAD only, the prevalence of PAD was 10.8% abetic persons. Estimating the prevalence of PAD is (95% CI 3.2%–18.4%) among those with difficult, because the majority of patients diabetes compared to 3.6% (95% CI 2.2%– Medical care utilization data also suggest are asymptomatic. Older studies used 5.0%) in those without diabetes (23). a higher frequency of PAD among persons claudication symptoms to identify those with diabetes, who had a fourfold higher with PAD, which underestimates the prev- The importance of using a sensitive occurrence of ambulatory care visits for alence. Data from physicians’ practices, in measure of PAD rather than one based PAD in the United States in 2002–2009 the Peripheral Artery Disease Awareness on symptoms can be seen by comparing compared to persons without diabetes Risk and Treatment Program (PARTNERS), PAD prevalence to the presence of inter- (Figure 20.2). The proportion of hospital- indicate that among participants with an mittent claudication (Table 20.3). Before izations listing PAD among the discharge ABI <0.9, 50% were asymptomatic, 40% discussing these results, it is important diagnoses was greater in persons with had claudication, and 10% had clinical to recognize that the presence of calf diabetes compared to those without lower extremity disease (21). Estimates pain while walking was used to suggest diabetes during 2002–2009 (Figure 20.3, of the prevalence of PAD are available presence of claudication and probably Table 20.4), regardless of age, sex, or race/ from the NHANES, where a modified overestimates the prevalence of true ethnicity categories examined. ABI test was performed. Instead of claudication due to PAD. Diabetes was measuring pressure in both the posterior associated with an approximately twofold Other correlates of higher prevalence of tibial and dorsalis pedis arteries as per higher prevalence of claudication in a PAD in the NHANES 1999–2004 were the usual recommendation, pressure new analysis of NHANES 1999–2004 greater age, non-Hispanic white or black was measured in the former location data. This association, though, was seen compared to all Hispanic race/ethnicity, only (22). New analysis of data from the in those with known diabetes. The preva- and former or current smoking (Table NHANES 1999–2004 showed that overall lence of claudication was similar among 20.2). In each age, sex, race/ethnicity, and prevalence of PAD, defined as an ABI nondiabetic persons compared to those smoking stratum, diabetes was associated <0.9 in either leg, in persons age ≥40 with undiagnosed diabetes (Table 20.3) with a higher prevalence of PAD (Figures years was 5.23% (Figure 20.1, Table 20.2). but higher in persons with diagnosed 20.4–20.7, Table 20.2).

20–4 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

Diabetes and PAD Risk Factors FIGURE 20.2. Percent of Outpatient Visits to a Physician Pertaining to Peripheral Arterial The Framingham Offspring Study exam- Disease, by Diabetes Status, U.S., 2002–2009 ined 1,554 males and 1,759 females Overall Diabetes No diabetes for PAD. In this population-based study, 2 the odds ratio for PAD was 2.3 (95% CI 1.5–3.6) for diabetic versus nondiabetic 1.5 participants (24). The Health Professionals Follow-up Study included 48,607 men 1 followed for 12 years (25). After adjusting for (CVD) risk 0.5 factors, the relative risk of developing PAD Peripheral arterial disease (%) 0 for men with diabetes compared with Diabetes Status men without diabetes was 2.61 (95% CI Peripheral arterial disease is defined based on ICD-9 codes 250.7, 440.2–440.4, 442.2, 442.3, 443.8, 443.9, 451.1, 1.98–3.45). and 451.2. Diabetes is defined based on ICD-9 codes 250, 357.2, 362.0, 366.41, 648.0, and 775.1. Error bars repre- sent 95% confidence intervals. ICD-9, International Classification of Diseases, Ninth Revision. Among patients who have , All relative standard errors ≤30% SOURCE: National Ambulatory Medical Care Surveys 2002–2009 PAD is more common than among the general population. In the Pittsburgh FIGURE 20.3. Percent of Hospital Discharges Listing Peripheral Arterial Disease, by Diabetes Epidemiology of Diabetes Complications Status, U.S., 2002–2009 Study of childhood-onset type 1 diabetes, women who had type 1 diabetes for 30 Overall Diabetes No diabetes years had a prevalence of PAD >30% 6 compared to only 11% for men when 5 determined by ABI <0.8 at rest or after 4 exercise (26). The Epidemiology of 3 Diabetes Interventions and Complications (EDIC) study, the long-term follow-up of 2 the Diabetes Control and Complications 1

Trial (DCCT), evaluated outcomes associ- Peripheral arterial disease (%) 0 ated with intensive versus conventional Diabetes Status glycemic control and identified those Peripheral arterial disease is defined based on ICD-9 codes 250.7, 440.2–440.4, 442.2, 442.3, 443.8, 443.9, 451.1, and patients with ABI <0.9. The EDIC study 451.2. Diabetes is defined based on ICD-9 codes 250, 357.2, 362.0, 366.41, 648.0, and 775.1. Error bars represent 95% confidence intervals. Confidence intervals were most likely underestimated because the National Hospital Discharge found that intensively treated partici- Survey sampling variables were not available, and consequently, it was not possible to take into account the complex pants, with an average duration of type 1 sampling design. See Table 20.4 for further details. ICD-9, International Classification of Diseases, Ninth Revision. All relative standard errors ≤30% diabetes of about 14 years, had a preva- SOURCE: National Hospital Discharge Surveys 2002–2009 lence of PAD of 8.8% among women and 4.6% among men (27). Although men have TABLE 20.4. Percent of Hospital Discharges Listing Peripheral Arterial Disease, Overall and a higher risk for by Age, Sex, Race, and Diabetes Status, U.S., 2002–2009 than women, PAD was shown in a compre- hensive to occur with PERCENT (STANDARD ERROR) equal frequency by sex in higher income CHARACTERISTICS Overall Diabetes No Diabetes countries and more frequently in women Overall 2.0 (0.01) 5.8 (0.07) 1.4 (0.01) in low to middle income countries (28). Age (years) <45 0.1 (0.05) 1.5 (0.09) 0.1 (0.004) 45–64 2.4 (0.03) 5.4 (0.11) 1.6 (0.03) Greater duration of diabetes is associated 65–74 4.3 (0.06) 7.3 (0.15) 3.4 (0.06) ≥75 3.9 (0.04) 6.6 (0.13) 3.2 (0.05) with a higher risk of developing PAD. Sex Compared with men without diabetes in Men 1.5 (0.02) 4.8 (0.08) 1.1 (0.02) the Health Professionals Follow-up Study, Women 2.6 (0.03) 7.0 (0.11) 1.8 (0.02) the relative risk for PAD was 1.39 (95% CI Race White 2.1 (0.02) 5.8 (0.10) 1.5 (0.02) 0.82–2.36) for 1–5 years of diabetes, 3.63 Black 2.0 (0.04) 6.1 (0.17) 1.1 (0.03) (95% CI 2.23–5.88) for 6–10 years, 2.55 AIAN 1.3 (0.16) 5.2 (0.76) 0.5 (0.11) (95% CI 1.50–4.32) for 11–25 years, and Asian 0.9 (0.08) 3.5 (0.49) 0.5 (0.07) 4.53 (95% CI 2.39–8.58) for >25 years (29). Peripheral arterial disease is defined based on ICD-9 codes 250.7, 440.2–440.4, 442.2, 442.3, 443.8, 443.9, 451.1, and 451.2. Diabetes is defined based on ICD-9 codes 250, 357.2, 362.0, 366.41, 648.0, and 775.1. Standard errors were most likely underestimated because the National Hospital Discharge Survey sampling variables were not Patients with in the United available, and consequently, it was not possible to take into account the complex sampling design. AIAN, American Indian/Alaska Native; ICD-9, International Classification of Diseases, Ninth Revision. Kingdom Prospective Diabetes Study All relative standard errors ≤30% (UKPDS) had a prevalence of PAD of SOURCE: National Hospital Discharge Surveys 2002–2009 20–5 DIABETES IN AMERICA, 3rd Edition

1.2% (95% CI 0.9%–1.5%) at the time of FIGURE 20.4. Prevalence of Peripheral Arterial Disease Among Adults Age ≥40 Years, by Age diagnosis of their diabetes (30). PAD in the and Diabetes Status, U.S., 1999–2004 UKPDS was defined as the presence of Overall All diabetes Diagnosed diabetes any two of the following: (1) ABI <0.8, Undiagnosed diabetes No diabetes (2) absence of both dorsalis pedis and 40 posterior tibial pulses to palpation in at 35 least one leg, and (3) claudication. At 6 30 25 years of follow-up in the UKPDS, 2.7% 20 of participants (95% CI 2.2%–3.2%) had 15 1 incident PAD according to these criteria, 10 1 and 10.6% had at least one of these 5

Peripheral arterial disease (%) 0 three abnormal measures. The preva- 4064 6574 ≥75 lence of PAD increased to 12.5% (95% Age (Years) CI 3.8%–21.1%) in a smaller subgroup of Peripheral arterial disease is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on participants followed for 18 years. In the self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions UKPDS, each 1% increase in glycosylated for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.2 for further details. A1c, glycosylated hemoglobin. hemoglobin (A1c) was associated with 1 Relative standard error >40%–50% a 28% (95% CI 12%–46%) increased risk SOURCE: National Health and Nutrition Examination Surveys 1999–2004 of PAD (30). The association with hyper- glycemia was independent of other risk FIGURE 20.5. Prevalence of Peripheral Arterial Disease Among Adults Age ≥40 Years, by Sex factors, including age, elevated systolic and Diabetes Status, U.S., 1999–2004 blood pressure, low high-density lipo- Overall All diabetes Diagnosed diabetes protein (HDL) , smoking, prior Undiagnosed diabetes No diabetes CVD, peripheral sensory neuropathy, and 20 (30). 18 16 14

A1c was measured in adults age ≥40 12 1 years in the NHANES 1999–2004 with 10 8 and without diabetes. According to a new 6 analysis for Diabetes in America, similar 4 mean A1c levels were seen by presence 2 Peripheral arterial disease (%) 0 of PAD among persons with diagnosed Men Women or undiagnosed diabetes (Table 20.5). Peripheral arterial disease is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on One reason for this similarity might be self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions that persons with PAD and diagnosed for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.2 for further details. A1c, glycosylated hemoglobin. diabetes were considered to be at higher 1 Relative standard error >30%–40% risk for complications and treated more SOURCE: National Health and Nutrition Examination Surveys 1999–2004 intensively than those with diabetes but without PAD. However, this assertion is FIGURE 20.6. Prevalence of Peripheral Arterial Disease Among Adults Age ≥40 Years, by not supported among persons with undi- Race/Ethnicity and Diabetes Status, U.S., 1999–2004 agnosed diabetes who would not have Overall All diabetes Diagnosed diabetes been targeted for diabetes treatment, Undiagnosed diabetes No diabetes where mean A1c was 0.51% lower among 20 persons with PAD (Table 20.5). Among persons without diabetes, mean A1c 15 was slightly higher in persons with PAD compared to those without, although the 10 mean A1c in persons with PAD would be classified as normal by American 5

Diabetes Association criteria (15). In 1 Peripheral arterial disease (%) 0 general, A1c differences by PAD presence Non-Hispanic White Non-Hispanic Blac Meican American among persons with and without diabetes Peripheral arterial disease is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on did not vary substantially when examined self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions within age, sex, and race/ethnicity strata for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.2 for further details. A1c, glycosylated hemoglobin. with the exception of the “all Hispanic” 1 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. All other relative standard errors ≤30%. group. Among all Hispanics, a difference SOURCE: National Health and Nutrition Examination Surveys 1999–2004

20–6 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

TABLE 20.5. Mean A1c (%) Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

MEAN (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 7.03 (0.16) 7.24 (0.07) 7.38 (0.21) 7.43 (0.09) 6.34 (0.10) 6.85 (0.12) 5.51 (0.03) 5.36 (0.01) Age (years) 40–64 7.46 (0.46) 7.44 (0.09) 8.07 (0.58) 7.69 (0.12) 6.50 (0.15) 6.91 (0.18) 5.53 (0.06) 5.33 (0.01) 65–74 6.84 (0.20) 6.98 (0.13) 7.16 (0.21) 7.09 (0.15) 6.05 (0.09) 6.76 (0.19) 5.48 (0.05) 5.45 (0.02) ≥75 6.83 (0.15) 6.71 (0.10) 7.02 (0.22) 6.72 (0.10) 6.47 (0.20) 6.68 (0.19) 5.49 (0.03) 5.49 (0.02) Sex Men 7.06 (0.22) 7.22 (0.11) 7.55 (0.30) 7.43 (0.14) 6.24 (0.10) 6.85 (0.15) 5.56 (0.04) 5.36 (0.02) Women 6.98 (0.21) 7.26 (0.10) 7.17 (0.26) 7.43 (0.11) 6.52 (0.22) 6.85 (0.16) 5.47 (0.04) 5.35 (0.02) Race/ethnicity Non-Hispanic white 6.84 (0.19) 6.94 (0.08) 7.20 (0.26) 7.09 (0.09) 6.22 (0.10) 6.66 (0.14) 5.49 (0.04) 5.34 (0.02) Non-Hispanic black 7.79 (0.28) 7.74 (0.12) 8.03 (0.26) 7.96 (0.16) 6.97 (0.25) 7.20 (0.27) 5.58 (0.06) 5.43 (0.02) All Hispanic 6.95 (0.48) 8.08 (0.20) 7.11 (0.67) 8.17 (0.27) 6.67 (0.38) 7.82 (0.24) 5.58 (0.08) 5.43 (0.02) Mexican American 7.70 (0.52) 8.01 (0.15) 7.76 (0.71) 8.13 (0.17) 7.50 (0.06) 7.74 (0.29) 5.45 (0.09) 5.41 (0.02) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. All relative standard errors ≤30% SOURCE: National Health and Nutrition Examination Surveys 1999–2004 of >1% in the A1c value was seen by FIGURE 20.7. Prevalence of Peripheral Arterial Disease Among Adults Age ≥40 Years, by presence of PAD for both known and Smoking Status and Diabetes Status, U.S., 1999–2004 undiagnosed diabetes, with lower values seen in those with PAD in these catego- Overall All diabetes Diagnosed diabetes Undiagnosed diabetes No diabetes ries (Table 20.5). 30 *2 25 Other Risk Factors for PAD Multiple factors other than diabetes 20 are associated with greater risk of PAD 15 1 including age, race/ethnicity, smoking, 10 , concentrations, inflam- 1 matory markers, and renal dysfunction. 5 Greater age and non-Hispanic black Peripheral arterial disease (%) 0 Never Smoers Former Smoers Current Smoers race are both associated with a higher prevalence of this condition in previous Peripheral arterial disease is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions publications and in the NHANES 1999– for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% 2004 data as discussed earlier (Figures confidence intervals. See Table 20.2 for further details. A1c, glycosylated hemoglobin. 20.4 and 20.6, Table 20.2) (23,24,31,32). * Upper confidence interval is 45.3%. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% The number of cigarettes smoked is SOURCE: National Health and Nutrition Examination Surveys 1999–2004 strongly associated with the incidence of intermittent claudication as demon- The greater prevalence of PAD among Hypertension was reported to be asso- strated in the Framingham Study, in smokers was seen in both the previously ciated with a threefold increased risk of which smoking was the strongest single undiagnosed and diagnosed diabetes intermittent claudication at the 16-year for development of symp- groups compared to persons without follow-up of the Framingham Study (34). tomatic PAD, regardless of sex (33). diabetes (Figure 20.7, Table 20.2). An The Cardiovascular Health Study reported Multiple studies reported that smoking overall higher prevalence of current or about a 50% higher prevalence of an ABI is associated with a twofold to fourfold former smoking was seen among persons <0.9 associated with hypertension in a increase in the risk of developing PAD with as opposed to those without PAD multivariate analysis adjusted for age, (24,30,34,35,36,37,38,39). In the in those with diagnosed or undiagnosed smoking, diabetes, and (32). NHANES 1999–2004 data, 15.74% of diabetes, as well as those without In a new analysis of the NHANES 1999– current smokers with diabetes had PAD diabetes (Figures 20.8 and 20.9, Tables 2004, among persons without diabetes, compared with 5.39% of never smokers 20.6 and 20.7). hypertension frequency was approximately with diabetes (Figure 20.7, Table 20.2). 70% higher in those with PAD compared to

20–7 DIABETES IN AMERICA, 3rd Edition those without PAD (Table 20.8). In these FIGURE 20.8. Prevalence of Current Smoking Among Adults Age ≥40 Years, by Peripheral data, the prevalence of hypertension in Arterial Disease and Diabetes Status, U.S., 1999–2004 diabetes was increased, so that persons with PAD had a higher prevalence of hyper- PAD No PAD 40 tension, but the elevation compared to 1 those without PAD was not as pronounced. 30 Women with diabetes, in particular, had a higher prevalence of hypertension than 20 men, with or without PAD. 10 The association of Current smokers (%) with atherosclerosis of the lower extremi- 0 ties has been known since the 1930s (40). All Diabetes Diagnosed Undiagnosed No Diabetes Diabetes Diabetes The prevalence of claudication in patients with serum cholesterol levels >260 mg/dL Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based (>6.73 mmol/L) is on average over twice on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% as high as in those with concentrations confidence intervals. See Table 20.6 for further details. A1c, glycosylated hemoglobin. below this level. The Edinburgh Artery * Upper confidence interval is 62.1%. 1 Relative standard error >30%–40% Study reported a higher prevalence of SOURCE: National Health and Nutrition Examination Surveys 1999–2004 PAD in association with higher serum cholesterol and lower HDL cholesterol in FIGURE 20.9. Prevalence of Former Smoking Among Adults Age ≥40 Years, by Peripheral multiple logistic regression analysis (41). Arterial Disease and Diabetes Status, U.S., 1999–2004 The Cardiovascular Health Study reached similar conclusions among its sample of PAD No PAD 5,084 subjects age ≥65 years, with PAD 80 defined as an ABI <0.9 (32). 60 A variety of novel risk factors have been associated with a higher prevalence of 40 PAD in several population-based studies. 20

Higher circulating levels of homocys- Former smokers (%) teine have been demonstrated with PAD 0 (42), as have low levels of folate in red All Diabetes Diagnosed Undiagnosed No Diabetes blood cells and circulating vitamin B6 Diabetes Diabetes (43). Higher levels of various hemostatic Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based factors have been demonstrated in on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions persons with low ABI, suggesting that a for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% hypercoagulable state predisposes to the confidence intervals. See Table 20.7 for further details. A1c, glycosylated hemoglobin. All relative standard errors ≤30% development of PAD (44,45). Increased SOURCE: National Health and Nutrition Examination Surveys 1999–2004 levels of hemostatic factors, such as fibrinogen, von Willebrand factor, tissue to those without preexisting disease and but not among persons with undiagnosed plasminogen activator (t-PA), D-dimer, a C-reactive protein of <1 mg/L (<9.52 diabetes, where mean C-reactive protein and plasma viscosity explained in part the nmol/L), those with diabetes and an concentration was lower (Table 20.9). higher prevalence of PAD in subjects with elevated C-reactive protein had an odds The reason for the lower mean C-reactive diabetes or impaired glucose tolerance in ratio for PAD of 8.6 (95% CI 2.2–34.0). protein in those with PAD among persons the Edinburgh Artery Study (46). Subjects with the with undiagnosed diabetes is unknown. and an elevated C-reactive protein also Mean C-reactive protein concentration A number of studies have shown an asso- had higher odds of PAD (odds ratio 3.9, was higher in persons with PAD among ciation between various inflammatory 95% CI 1.1–14.6). A new analysis of those diagnosed with diabetes when markers and PAD. C-reactive protein NHANES 1999–2004 data showed that subjects were further stratified by age, sex, and the presence of PAD, defined as ABI among participants age ≥40 years, mean or race/ethnicity, with the exception of <0.9, were studied among 1,600 subjects C-reactive protein concentration was non-Hispanic black subjects (Table 20.9). with the metabolic syndrome, diabetes, higher among those with PAD compared An earlier report of the NHANES 1999– or preexisting arterial disease in the to those without PAD among persons with 2002 found that inflammatory markers, NHANES 1999–2000 (47). Compared diagnosed diabetes and without diabetes including C-reactive protein, fibrinogen,

20–8 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

TABLE 20.6. Prevalence of Current Smoking Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 26.7 (5.1) 17.7 (1.5) 22.1 (3.7) 19.2 (1.7) 36.0 (12.9)1 14.3 (2.6) 27.8 (4.0) 20.5 (1.5) Age (years) 40–64 54.7 (8.2) 23.7 (2.0) 48.5 (11.0) 25.4 (2.2) 64.5 (20.2)1 20.1 (3.9) 39.6 (8.2) 23.8 (1.7) 65–74 20.6 (7.6)1 8.1 (1.8) 13.9 (5.7)2 9.5 (2.1) 3 3 29.5 (8.3) 10.5 (1.8) ≥75 3 5.1 (2.3)2 3 3 3 3 7.9 (3.7) 2 4.7 (1.4) Sex Men 30.3 (6.3) 19.7 (2.1) 27.2 (5.7) 21.9 (2.6) 35.4 (16.4)2 15.7 (4.1) 37.1 (7.0) 22.9 (1.9) Women 22.0 (7.8)1 15.2 (2.2) 16.1 (6.0)1 16.4 (2.7) 3 12.1 (4.1)1 22.2 (5.5) 18.5 (1.5) Race/ethnicity Non-Hispanic white 26.7 (7.1) 14.2 (2.1) 19.0 (5.0) 16.5 (2.2) 40.3 (15.6)1 9.7 (2.9) 27.6 (4.6) 20.1 (1.7) Non-Hispanic black 34.3 (5.8) 26.8 (3.3) 35.8 (6.4) 25.9 (3.9) 3 29.1 (7.2) 21.5 (7.1) 1 25.7 (2.8) All Hispanic 10.0 (4.1)2 21.6 (3.4) 15.6 (4.3) 21.4 (4.2) 3 22.0 (6.2) 3 24.0 (3.3) Mexican American 14.2 (6.7)2 21.0 (2.0) 18.7 (7.7)2 19.6 (2.7) 3 24.4 (4.7) 18.0 (6.2)1 20.8 (2.7) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

TABLE 20.7. Prevalence of Former Smoking Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 52.1 (6.4) 36.5 (2.1) 53.6 (6.2) 34.5 (1.9) 49.0 (12.4) 40.7 (4.2) 38.6 (5.1) 32.5 (1.2) Age (years) 40–64 27.7 (8.5)1 31.3 (2.5) 22.8 (7.6) 1 31.0 (2.3) 3 31.9 (5.0) 26.6 (7.4) 29.5 (1.4) 65–74 63.7 (8.9) 47.5 (3.4) 66.1 (8.7) 40.4 (4.0) 57.4 (22.3) 1 61.7 (5.0) 44.3 (8.3) 43.7 (2.7) ≥75 62.1 (10.0) 41.4 (4.9) 65.5 (10.7) 41.5 (4.8) 55.7 (15.9) 41.3 (10.0) 51.0 (8.0) 43.3 (3.4) Sex Men 48.9 (8.3) 45.1 (2.8) 46.9 (7.7) 44.3 (2.9) 52.4 (16.0)1 46.7 (5.5) 52.7 (6.9) 38.0 (1.7) Women 56.3 (9.4) 26.0 (2.6) 61.5 (9.0) 23.9 (2.8) 43.1 (19.9)2 31.4 (6.1) 30.2 (5.9) 27.5 (1.5) Race/ethnicity Non-Hispanic white 52.9 (8.4) 42.7 (3.3) 56.5 (8.3) 39.8 (3.1) 46.6 (14.8)1 48.2 (5.1) 41.5 (6.2) 34.5 (1.4) Non-Hispanic black 43.3 (6.1) 26.9 (2.8) 45.9 (5.9) 27.4 (2.9) 3 25.3 (7.6) 32.7 (8.7) 21.2 (2.1) All Hispanic 72.7 (7.6) 25.5 (3.1) 57.1 (12.6) 26.0 (4.0) 3 24.1 (5.2) 24.5 (11.8)2 25.5 (2.1) Mexican American 53.7 (6.7) 34.1 (3.3) 39.1 (8.8) 34.2 (4.8) 3 33.7 (5.5) 53.7 (10.3) 23.5 (2.0) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004 and leukocyte count, were independently Other factors associated with PAD have renal insufficiency, defined as a reduced associated with PAD among 4,787 partic- been reported using NHANES data. Blood estimated glomerular filtration rate (eGFR ipants age ≥40 years (48). The InCHIANTI cadmium levels were associated with <60 mL/min/1.73 m2), was associated study, a population-based Italian study increased prevalence of PAD, as defined with an odds ratio of 2.17 (95% CI 1.10– that enrolled 955 men and women age by an ABI <0.9, in the NHANES 1999– 4.30) for prevalent PAD (23). This analysis ≥60 years, found that subjects with PAD 2000 among subjects age ≥40 years (50). adjusted for diabetes, coronary artery had higher levels of interleukin (IL)-1 The highest quartile of cadmium level disease, , hypertension, body mass receptor antagonist, IL-6, fibrinogen, and compared to the lowest was associated index (BMI), total cholesterol, diastolic and C-reactive protein compared to subjects with an odds ratio for PAD of 2.82 (95% CI systolic blood pressures, and smoking without PAD (49). 1.36–5.85). An analysis of the NHANES history. Other studies have shown similar 1999–2000 population demonstrated that results (30,36,51,52,53).

20–9 DIABETES IN AMERICA, 3rd Edition

TABLE 20.8. Prevalence of Hypertension Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 75.9 (4.93) 65.3 (1.74) 77.8 (4.17) 64.2 (2.07) 71.7 (12.59) 67.8 (3.11) 66.2 (5.17) 38.9 (1.37) Age (years) 40–64 63.5 (9.55) 58.3 (2.89) 74.0 (9.46) 53.9 (3.05) 1 67.6 (4.40) 50.7 (8.50) 31.5 (1.57) 65–74 84.2 (5.53) 76.2 (3.54) 81.3 (6.73) 81.4 (3.41) 93.6 (6.81) 66.2 (7.87) 78.5 (6.57) 61.1 (2.61) ≥75 78.3 (6.67) 79.8 (3.93) 76.2 (7.89) 83.3 (3.65) 82.4 (15.21) 72.3 (8.04) 75.9 (6.12) 75.7 (2.44) Sex Men 71.8 (7.72) 59.1 (2.52) 75.4 (5.89) 56.2 (2.85) 65.9 (17.27) 64.1 (4.51) 69.8 (6.27) 36.2 (1.93) Women 81.5 (4.58) 72.9 (2.57) 80.5 (5.84) 72.7 (2.87) 84.6 (7.70) 73.4 (4.67) 63.9 (5.98) 41.3 (1.82) Race/ethnicity Non-Hispanic white 73.5 (6.21) 67.4 (2.32) 75.9 (5.27) 65.2 (3.17) 68.8 (15.21) 71.6 (4.03) 64.3 (6.05) 38.8 (1.63) Non-Hispanic black 80.7 (5.49) 74.5 (2.95) 83.0 (6.35) 77.2 (3.96) 71.8 (14.94) 67.6 (7.03) 72.3 (9.48) 51.7 (2.73) All Hispanic 88.2 (4.28) 51.2 (4.84) 81.5 (6.90) 52.2 (6.06) 1 48.3 (8.41) 80.2 (8.59) 30.2 (2.74) Mexican American 82.3 (4.97) 52.3 (3.12) 76.7 (8.17) 54.2 (4.18) 1 47.4 (6.96) 64.1 (13.03) 30.9 (2.25) Hypertension is defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or use of antihypertensive . Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

TABLE 20.9. Mean C-Reactive Protein (mg/L) Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

MEAN (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 0.86 (0.20) 0.57 (0.03) 1.02 (0.27) 0.55 (0.04) 0.54 (0.17)1 0.61 (0.07) 0.69 (0.07) 0.42 (0.02) Age (years) 40–64 1.14 (0.55)2 0.62 (0.04) 3 0.58 (0.05) 0.39 (0.11) 0.71 (0.11) 0.79 (0.11) 0.41 (0.02) 65–74 0.67 (0.18) 0.45 (0.04) 0.64 (0.15) 0.48 (0.05) 3 0.39 (0.04) 0.78 (0.21) 0.51 (0.05) ≥75 0.79 (0.09) 0.56 (0.11) 0.98 (0.18) 0.57 (0.13) 0.46 (0.07) 0.52 (0.15) 0.44 (0.07) 0.44 (0.04) Sex Men 0.80 (0.32)2 0.44 (0.03) 1.07 (0.45)2 0.44 (0.05) 0.34 (0.08) 0.44 (0.08) 0.71 (0.16) 0.35 (0.02) Women 0.94 (0.15) 0.73 (0.05) 0.96 (0.17) 0.68 (0.06) 0.88 (0.36)2 0.87 (0.13) 0.68 (0.07) 0.49 (0.03) Race/ethnicity Non-Hispanic white 0.87 (0.27)1 0.53 (0.04) 1.10 (0.38)1 0.54 (0.04) 0.49 (0.20)2 0.50 (0.07) 0.68 (0.09) 0.42 (0.02) Non-Hispanic black 0.77 (0.14) 0.91 (0.14) 0.75 (0.15) 0.75 (0.08) 0.84 (0.22) 1.30 (0.38) 0.82 (0.25) 0.54 (0.05) All Hispanic 0.97 (0.33)1 0.57 (0.05) 1.21 (0.59)2 0.54 (0.06) 0.54 (0.07) 0.66 (0.09) 0.61 (0.16) 0.42 (0.02) Mexican American 1.29 (0.48)1 0.57 (0.04) 3 0.54 (0.04) 0.58 (0.19)1 0.63 (0.09) 0.45 (0.11) 0.45 (0.03) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

In NHANES 1999–2004 data analyzed for as a consistent accompanying feature 1999–2004 confirms a higher mean Diabetes in America, mean serum creat- of PAD among persons with diabetes urinary albumin-to-creatinine ratio among inine, a marker of renal function, was according to these data. Presence of persons with PAD for those with known higher in subjects with PAD compared to albumin in the urine also reflects renal diabetes and for men without diabetes those without PAD regardless of diabetes dysfunction, and a spot urine albu- (Figure 20.11, Table 20.11). Estimates of status (Figure 20.10, Table 20.10). The min-to-creatinine ratio exceeding the the mean among women lacked sufficient same association was found when data threshold for microalbuminuria was precision to permit similar comparisons were stratified by age, sex, and race/ associated with a higher prevalence of (Table 20.11). A published analysis of the ethnicity for both subjects with known an ABI >0.9 in the Multi-Ethnic Study NHANES 1999–2004 of PAD prevalence and undiagnosed diabetes (Table 20.10). of Atherosclerosis (MESA) study (54). in relation to eGFR <60 mL/min/1.73 m2 Higher elevated serum creatinine appears A new analysis of NHANES data from and microalbuminuria, defined as urinary

20–10 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes albumin-to-creatinine ratio >30 mg/g, FIGURE 20.10. Mean Serum Creatinine (mg/dL) Among Adults Age ≥40 Years, by found the following odds ratios for the Peripheral Arterial Disease and Diabetes Status, U.S., 1999–2004 dimensions of renal function singly and PAD No PAD in combination: microalbuminuria, 1.72 1.5 (95% CI 1.16–2.55); eGFR, 1.58 (95% CI 1.09–2.29); and both microalbuminuria 1 and eGFR, 2.26 (95% CI 1.30–3.94) (55). NHANES 1999–2004 data showed that urinary albumin-to-creatinine ratio ≥30 0.5 mg/g was associated with PAD among persons age 40–64 years and in men Serum creatinine (mgdL) All Diabetes Diagnosed Undiagnosed No Diabetes only (Figure 20.12, Table 20.12). The Diabetes Diabetes association between PAD and renal Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based dysfunction is not entirely understood on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions and is explained only in part by shared for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.10 for further details. A1c, glycosylated hemoglobin. risk factors (56). All relative standard errors ≤30% SOURCE: National Health and Nutrition Examination Surveys 1999–2004 According to new analyses conducted for Diabetes in America, persons with diabetes FIGURE 20.11. Mean Urinary Albumin-to-Creatinine Ratio (mg/g) Among Adults Age ≥40 are more likely to have generalized and Years, by Peripheral Arterial Disease and Diabetes Status, U.S., 1999–2004 abdominal adiposity than those without PAD No PAD diabetes, as seen in the higher mean BMI 600 and waist circumference measurements 500 in the NHANES 1999–2004 population 400 by diabetes status (Appendices 20.1 and 1 300 20.2). The same direction of association 1 is not seen within the diabetes categories 200 by PAD presence. Instead mean BMI 100 2 3 and waist circumference are similar or

Albumin-to-creatinine ratio (mgg) All Diabetes Diagnosed Undiagnosed No Diabetes slightly lower among persons with PAD Diabetes Diabetes (Appendices 20.1 and 20.2). Potential Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based explanations for this association are that on self-report. Undiagnosed diabetes is defined by A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions adiposity is not related to risk of PAD for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.11 for further details. A1c, glycosylated hemoglobin. among persons with diabetes or that 1 Relative standard error >30%–40% adiposity is lower among persons with 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. PAD due to higher smoking prevalence. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

TABLE 20.10. Mean Serum Creatinine (mg/dL) Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

MEAN (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 1.14 (0.05) 0.90 (0.01) 1.15 (0.06) 0.91 (0.02) 1.12 (0.10) 0.89 (0.02) 1.01 (0.08) 0.86 (0.01) Age (years) 40–64 0.99 (0.05) 0.85 (0.02) 1.08 (0.08) 0.86 (0.02) 0.86 (0.04) 0.83 (0.03) 0.81 (0.03) 0.85 (0.01) 65–74 1.13 (0.08) 0.97 (0.03) 1.10 (0.08) 0.98 (0.03) 1.19 (0.21) 0.93 (0.03) 1.21 (0.25) 0.87 (0.01) ≥75 1.31 (0.07) 1.04 (0.04) 1.29 (0.09) 1.01 (0.02) 1.34 (0.11) 1.10 (0.11) 1.12 (0.06) 0.97 (0.02) Sex Men 1.22 (0.08) 0.99 (0.02) 1.24 (0.08) 1.00 (0.02) 1.19 (0.15) 0.97 (0.03) 1.07 (0.05) 0.99 (0.02) Women 1.01 (0.06) 0.80 (0.02) 1.02 (0.08) 0.81 (0.02) 0.98 (0.08) 0.75 (0.03) 0.98 (0.13) 0.74 (0.01) Race/ethnicity Non-Hispanic white 1.13 (0.06) 0.90 (0.01) 1.14 (0.08) 0.90 (0.02) 1.11 (0.12) 0.90 (0.03) 1.05 (0.11) 0.87 (0.01) Non-Hispanic black 1.19 (0.06) 1.06 (0.07) 1.22 (0.08) 1.09 (0.09) 1.06 (0.05) 0.99 (0.11) 0.85 (0.05) 0.92 (0.02) All Hispanic 1.08 (0.15) 0.79 (0.03) 0.94 (0.10) 0.81 (0.03) 1.31 (0.24) 0.74 (0.04) 0.82 (0.07) 0.75 (0.02) Mexican American 0.89 (0.13) 0.79 (0.02) 0.92 (0.16) 0.80 (0.03) 0.79 (0.04) 0.77 (0.04) 0.93 (0.09) 0.76 (0.02) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. All relative standard errors ≤30% SOURCE: National Health and Nutrition Examination Surveys 1999–2004

20–11 DIABETES IN AMERICA, 3rd Edition

TABLE 20.11. Mean Urinary Albumin-to-Creatinine Ratio (mg/g) Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

MEAN (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 244.9 (83.90)1 87.7 (11.97) 340.6 (128.07)1 102.5 (15.05) 54.8 (23.12)2 56.4 (13.80) 3 16.3 (1.45) Age (years) 40–64 3 83.4 (16.26) 3 91.5 (18.69) 21.5 (9.69)2 66.0 (21.72)1 17.7 (5.37) 1 14.4 (1.75) 65–74 3 95.2 (20.72) 3 128.6 (31.32) 3 30.4 (6.93) 3 18.7 (3.52) ≥75 127.2 (46.13)1 95.1 (31.94)1 172.3 (74.99) 2 109.6 (46.14)2 40.9 (20.37)2 62.9 (16.21) 54.8 (17.74)1 31.9 (4.39) Sex Men 360.5 (143.53)1 106.4 (20.08) 549.1 (226.78)2 126.1 (25.29) 3 70.5 (22.49)1 59.6 (18.81)1 18.2 (2.80) Women 86.3 (25.56) 65.0 (9.49) 88.5 (28.24)1 76.8 (13.39) 3 34.6 (6.99) 3 14.6 (1.47) Race/ethnicity Non-Hispanic white 212.1 (98.79)2 62.1 (11.01) 3 75.5 (15.94) 51.3 (25.50)2 36.6 (4.97) 3 14.6 (1.53) Non-Hispanic black 3 164.7 (43.61) 3 179.3 (50.43) 16.0 (2.09) 3 36.0 (15.19)2 33.8 (8.10) All Hispanic 265.6 (98.57)1 167.0 (42.99) 330.2 (105.47)1 176.6 (48.31) 3 3 31.7 (10.10)1 14.7 (1.94) Mexican American 234.3 (71.54)1 201.3 (55.66) 3 203.7 (49.19) 417.4 (151.71)1 3 51.0 (15.69)1 18.5 (3.15) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

Diabetes Treatment and PAD Risk FIGURE 20.12. Prevalence of Urinary Albumin-to-Creatinine Ratio ≥30 mg/g Among Adults In the EDIC study, intensive insulin Age ≥40 Years, by Peripheral Arterial Disease and Diabetes Status, U.S., 1999–2004 therapy compared to conventional PAD No PAD therapy during the DCCT resulted in 60 decreased progression of carotid artery 50 intima-media thickness 6 years after the end of the trial (57). Progression 40 of carotid intima-media thickness was 30 1 associated with the traditional risk factors 20 mentioned above for PAD, including age, ≥30 mg/g (%) 10

systolic blood pressure, smoking, the Albumin-to-creatinine ratio ratio of low-density to HDL All Diabetes Diagnosed Undiagnosed No Diabetes cholesterol, urinary albumin excretion Diabetes Diabetes rate, and mean A1c value during the Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based DCCT. The A1c value explained 96% on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions of the differences between treatment for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.12 for further details. A1c, glycosylated hemoglobin. groups in intima-media thickness of 1 Relative standard error >40%–50% the at year 6 SOURCE: National Health and Nutrition Examination Surveys 1999–2004 of follow-up. These findings argue that progression of atherosclerosis can be prevalence of PAD, as insulin is often used Outcomes of PAD impeded with intensive glycemic control. for intensive control. In standard medical PAD is a progressive condition that exacts Whether these results would apply as practice, insulin use may be a marker for a substantial toll in terms of morbidity and well to the peripheral arteries of the lower more severe diabetes, and the associa- need for medical interventions. Estimates extremities is not known. tion between insulin use and higher PAD derived from population data show that prevalence may be another example of approximately 27% of patients with PAD Persons with diagnosed diabetes and confounding by indication. The use of oral experience progression of symptoms PAD included in the NHANES 1999–2004 for diabetes treatment was over a 5-year period (1). In a study of 257 more frequently reported insulin use slightly less common in persons with PAD patients with intermittent claudication (Table 20.13) for all examined age, sex, and diagnosed diabetes (Table 20.14). referred to a Copenhagen hospital-based and race/ethnicity categories in analyses This is not a surprising finding given that physiology clinic for initial evaluation, the conducted for Diabetes in America. This the same persons were more likely to be rate of clinical progression to rest pain or finding should not be interpreted to imply treated with insulin, which often replaces gangrene was 7.5% in the first year after that insulin is associated with a higher oral diabetes treatments. initial referral and 2.2% per year after that

20–12 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

TABLE 20.12. Prevalence of Urinary Albumin-to-Creatinine Ratio ≥30 mg/g Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/ Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 34.9 (4.9) 27.8 (1.6) 41.8 (5.3) 29.0 (1.9) 21.0 (9.0)2 25.5 (3.5) 21.7 (4.3) 7.3 (0.5) Age (years) 40–64 45.2 (7.9) 26.0 (2.3) 64.3 (9.8) 26.6 (2.5) 3 24.7 (5.0) 8.9 (4.2)2 5.7 (0.6) 65–74 29.8 (8.3) 32.1 (3.1) 32.0 (7.6) 33.6 (3.8) 3 29.3 (6.4) 31.5 (7.9) 10.6 (1.7) ≥75 30.6 (5.5) 28.7 (3.1) 33.8 (8.4) 32.3 (4.2) 24.5 (10.5)2 20.9 (5.7) 31.0 (6.4) 18.1 (1.9) Sex Men 40.5 (7.3) 31.0 (2.5) 57.5 (7.9) 32.9 (2.6) 3 27.5 (5.0) 25.4 (5.1) 6.2 (0.6) Women 27.1 (7.6) 24.0 (2.5) 22.9 (5.8) 24.7 (2.8) 3 22.3 (4.9) 19.5 (5.5) 8.3 (0.8) Race/ethnicity Non-Hispanic white 32.0 (6.5) 24.9 (1.9) 36.1 (7.1) 24.9 (2.0) 25.0 (11.6)2 25.1 (4.5) 22.0 (5.2) 6.4 (0.6) Non-Hispanic black 44.4 (8.3) 31.3 (3.6) 55.8 (7.6) 35.8 (4.1) 3 19.3 (5.3) 24.6 (9.6)1 12.5 (1.4) All Hispanic 44.6 (15.6)1 37.4 (4.3) 65.6 (9.7) 37.3 (4.7) 3 37.9 (8.4) 3 7.3 (1.5) Mexican American 44.9 (9.1) 32.2 (3.1) 52.2 (7.4) 34.2 (3.5) 3 27.5 (5.8) 18.5 (5.9)1 9.2 (1.3) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

(58). In this study, the cumulative rate of TABLE 20.13. Prevalence of Insulin Use Among Adults Age ≥40 Years With Diagnosed Diabetes, reconstructive lower extremity surgery Overall and by Age, Sex, Race/Ethnicity, and Peripheral Arterial Disease, U.S., 1999–2004 at 5 years was 9.5%, and the cumulative PERCENT (STANDARD ERROR) amputation rate was 6.8%. In a new CHARACTERISTICS PAD No PAD analysis of cross-sectional data from Overall 38.6 (6.2) 20.0 (2.1) the NHANES 1999–2004 population, a Age (years) greater proportion of persons with PAD 40–64 32.5 (10.8)1 18.2 (2.7) reported fair or poor health regardless 65–74 40.9 (7.3) 26.2 (3.9) of diabetes status (Table 20.15). There ≥75 41.1 (9.6) 17.0 (3.6) were too few persons with undiagnosed Sex Men 33.8 (6.4) 18.3 (2.7) diabetes and PAD to produce stable Women 44.3 (9.6) 21.8 (2.9) estimates by age, sex, and race/ethnicity Race/ethnicity strata; although in the strata in which such Non-Hispanic white 36.7 (7.9) 19.2 (3.3) estimates were available, the same trend Non-Hispanic black 49.3 (7.6) 26.7 (3.2) All Hispanic 16.7 (5.9)1 16.7 (2.1) was seen with poorer health reported by Mexican American 25.9 (7.8) 1 20.0 (2.6) persons with PAD (Table 20.15). Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Among those with diagnosed diabetes, 64% of participants with PAD and 71% of participants without PAD used oral diabetes medications. Cardiovascular Outcomes 1 Relative standard error >30%–40% Patients with symptomatic PAD have SOURCE: National Health and Nutrition Examination Surveys 1999–2004 four to seven times the risk of mortality from all causes and a fifteenfold higher is a hallmark of generalized atheroscle- myocardial regardless of risk of mortality from CVD than persons rosis and, therefore, would be expected diabetes status and in all strata defined who do not have PAD (59). Mortality to confer an increased risk of cardiovas- by age, sex, or race/ethnicity with rates appear to be related to the severity cular or cerebrovascular death, severe enough subjects to permit stable esti- of the obstructive process as measured PAD appears to carry a particularly mates (Figure 20.13, Table 20.16). These by the ABI. In prospective studies, PAD ominous prognosis. Patients with an ABI data support the coexistence of vascular mortality outcome by diabetes status is ≤0.30 had a very high 6-year cumulative arterial disease in the lower extremities not available, although a 6-year study mortality rate (64%) (60). Among the with the same disease process in the showed that low ABI was strongly NHANES 1999–2004 population, a new . associated with increased mortality, inde- data analysis showed that presence of pendent of age or presence or absence PAD was associated with a substantially Those with PAD severe enough to warrant of diabetes (60). Although the presence higher overall frequency of past history revascularization have particularly poor of arterial obstructive disease of the legs of coronary disease, , or outcomes. One study of outcomes in

20–13 DIABETES IN AMERICA, 3rd Edition revascularization found the cumulative is associated with disease elsewhere cardiovascular risk factors, was 2.0 (37). 6-year mortality rate was 62% in patients in the arterial system. The Honolulu The Atherosclerosis Risk in Communities with symptoms sufficiently severe to Heart Program enrolled 8,006 men of (ARIC) Study enrolled 15,792 people age require femoropopliteal bypass (61), while Japanese ancestry age 45–68 years 45–64 years and followed them for 7 another showed that 48% of patients with without known atherosclerosis, living years. Those with the lowest ABI also had claudication, 80% of those with ischemic on Oahu, Hawaii, and followed them approximately double the risk of stroke rest pain, and 95% of those with gangrene for 3–6 years. The risk of stroke asso- (39). As with the association between died within 10 years of undergoing ciated with an ABI <0.9, adjusted for PAD and coronary heart disease in the femoropopliteal bypass grafting (62). A university-based vascular surgery clinic TABLE 20.14. Prevalence of Oral Use Among Adults Age ≥40 Years in the Netherlands prospectively studied With Diagnosed Diabetes, Overall and by Age, Sex, Race/Ethnicity, and Peripheral Arterial 3,209 patients for an average of 8 years Disease, U.S., 1999–2004 and found that resting and post-exercise PERCENT (STANDARD ERROR) ABI values were strong and independent predictors of mortality (53). Mortality CHARACTERISTICS PAD No PAD increased by 8% for every 0.1 decrease Overall 64.3 (6.4) 70.7 (2.2) in resting ABI and by 9% for every 0.1 Age (years) decrease in post-exercise ABI. Among 40–64 62.9 (11.4) 72.7 (2.8) 65–74 63.5 (9.3) 68.2 (4.4) those who began the study with a normal ≥75 67.1 (10.4) 64.8 (4.8) ABI, a reduction in the post-exercise ABI Sex by 6%–24% was associated with a 1.6-fold Men 64.5 (8.9) 71.5 (3.0) increased risk of mortality; those with Women 64.1 (6.7) 69.8 (3.0) a reduction of 25%–55% had a 3.5-fold Race/ethnicity increase in mortality; and those with Non-Hispanic white 66.4 (8.1) 69.8 (3.4) a >55% reduction in ABI had a 4.8-fold Non-Hispanic black 64.8 (6.1) 69.9 (3.0) All Hispanic 48.1 (11.5) 70.6 (4.9) increase in mortality. Mexican American 69.0 (7.7) 79.0 (2.8) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based The risk of stroke is approximately on self-report. Among those with diagnosed diabetes, 39% of participants with PAD and 20% of participants without doubled among those with an ABI <0.9, PAD used insulin. All relative standard errors ≤30% indicating that the presence of PAD SOURCE: National Health and Nutrition Examination Surveys 1999–2004

TABLE 20.15. Prevalence of Fair or Poor Self-Reported Health Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 48.5 (5.6) 35.4 (2.4) 57.1 (5.8) 40.7 (2.6) 31.3 (11.7)1 24.1 (3.2) 33.4 (5.3) 17.1 (1.0) Age (years) 40–64 49.3 (12.4) 36.6 (3.2) 60.6 (11.6) 41.8 (3.5) 3 25.4 (4.7) 34.3 (8.8) 16.2 (1.1) 65–74 57.9 (8.1) 31.4 (3.3) 68.4 (6.5) 35.9 (3.7) 3 22.5 (6.1) 35.7 (7.9) 18.7 (2.1) ≥75 34.5 (8.5) 36.9 (4.6) 36.1 (9.6) 44.3 (4.8) 31.7 (12.6)1 20.3 (7.9) 1 29.5 (6.6) 22.8 (2.3) Sex Men 40.6 (7.4) 29.9 (2.8) 50.0 (7.6) 34.9 (3.3) 3 20.9 (3.8) 40.1 (7.4) 16.1 (1.2) Women 59.1 (9.5) 42.0 (2.6) 65.4 (9.6) 47.0 (3.2) 43.1 (18.9)2 28.9 (5.6) 29.3 (6.7) 18.0 (1.4) Race/ethnicity Non-Hispanic white 42.2 (6.4) 31.0 (3.4) 51.5 (7.2) 36.5 (4.0) 25.8 (12.6)2 20.7 (3.9) 31.0 (5.9) 14.8 (1.3) Non-Hispanic black 69.4 (6.6) 44.3 (3.3) 68.1 (7.8) 49.2 (3.6) 73.9 (16.0) 31.0 (8.2) 31.9 (8.9) 21.7 (2.8) All Hispanic 57.0 (17.0) 52.6 (4.4) 75.2 (7.7) 55.5 (6.0) 3 44.5 (7.8) 62.0 (9.1) 31.7 (2.9) Mexican American 63.6 (13.7) 57.3 (2.7) 61.6 (11.8) 60.5 (3.0) 69.8 (27.2)1 49.8 (5.9) 46.8 (10.5) 30.8 (2.5) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

20–14 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

NHANES 1999–2004, the same positive FIGURE 20.13. Percent of Adults Age ≥40 Years Previously Diagnosed With Coronary association is seen between PAD and Heart Disease, Angina, or , by Peripheral Arterial Disease and a previous diagnosis of stroke among Diabetes Status, U.S., 1999–2004 persons with and without diabetes in PAD No PAD cells with sufficient sample size to permit 75 a stable estimate, further affirming the coexistence of arterial vascular disease 60 in other beds in those with disease 45 affecting the lower extremities (Table 20.17). 30

15 or myocardial infarction (%)

Coronary heart disease, angina, All Diabetes Diagnosed Undiagnosed No Diabetes Diabetes Diabetes

Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Previous diagnosis of coronary heart disease, angina, or myocardial infarction is based on self-report. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. Error bars represent 95% confidence intervals. See Table 20.16 for further details. A1c, glycosylated hemoglobin. All relative standard errors ≤30% SOURCE: National Health and Nutrition Examination Surveys 1999–2004

TABLE 20.16. Percent of Adults Age ≥40 Years Previously Diagnosed With Coronary Heart Disease, Angina, or Myocardial Infarction, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 41.8 (6.1) 17.7 (2.0) 39.3 (5.6) 19.5 (2.4) 46.8 (12.4) 13.8 (2.3) 22.0 (4.4) 7.0 (0.5) Age (years) 40–64 42.0 (14.8)1 11.9 (1.8) 30.0 (10.4)1 12.6 (2.2) 60.9 (20.9)1 10.5 (2.9) 16.7 (8.2)2 4.4 (0.6) 65–74 41.0 (8.5) 26.1 (4.6) 37.9 (7.9) 28.3 (5.3) 49.0 (23.0)2 21.9 (5.7) 24.0 (6.9) 14.9 (1.7) ≥75 42.8 (9.5) 31.3 (3.6) 51.7 (9.0) 39.4 (4.2) 3 13.6 (4.0) 27.9 (6.9) 20.8 (2.1) Sex Men 52.2 (9.2) 20.1 (2.7) 44.8 (8.1) 22.2 (3.5) 64.6 (13.9) 16.3 (3.4) 37.5 (8.4) 9.3 (0.8) Women 27.6 (7.5) 14.7 (1.8) 32.5 (8.4) 16.6 (2.1) 3 9.8 (3.4)1 12.5 (4.0)1 5.1 (0.7) Race/ethnicity Non-Hispanic white 47.7 (7.6) 20.2 (2.4) 45.3 (6.6) 23.0 (3.1) 51.8 (15.1) 15.0 (2.7) 22.5 (5.1) 7.6 (0.6) Non-Hispanic black 28.1 (5.4) 16.0 (2.2) 26.1 (6.6) 14.0 (2.8) 35.6 (16.9)2 21.4 (6.1) 3 4.5 (1.0) All Hispanic 3 8.0 (1.7) 3 9.1 (2.1) 3 4.9 (2.4)2 34.8 (15.2)2 4.6 (1.5)1 Mexican American 3 10.9 (2.3) 12.4 (6.2)2 12.2 (2.3) 3 8.0 (3.9)2 27.8 (11.2)2 3.5 (0.7) Previous diagnoses of coronary heart disease, angina, and myocardial infarction are based on self-report. Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

20–15 DIABETES IN AMERICA, 3rd Edition

TABLE 20.17. Percent of Adults Age ≥40 Years Previously Diagnosed With Stroke, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

PERCENT (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 17.2 (4.2) 5.8 (1.0) 19.9 (5.6) 6.2 (1.0) 3 5.0 (1.8)1 8.3 (2.8)1 2.8 (0.4) Age (years) 40–64 16.4 (4.3) 4.0 (1.1) 3 3.8 (1.2)1 3 3 3 1.7 (0.4) 65–74 21.1 (7.5) 1 8.4 (1.7) 20.2 (6.8)1 10.5 (2.1) 3 3 17.6 (7.3) 2 4.8 (1.2) ≥75 12.5 (4.6)1 10.4 (3.0) 17.3 (7.2) 2 10.7 (3.2) 3 9.5 (4.4)2 9.1 (3.7)2 10.7 (1.4) Sex Men 16.5 (3.9) 5.6 (1.0) 23.7 (7.0) 4.9 (0.9) 3 6.8 (2.8)2 12.2 (4.5)1 2.1 (0.4) Women 18.2 (8.0)2 6.2 (1.5) 15.5 (6.4)2 7.7 (1.8) 3 3 6.0 (2.9)2 3.5 (0.5) Race/ethnicity Non-Hispanic white 15.1 (5.4)1 6.5 (1.6) 18.0 (7.2)1 6.8 (1.5) 3 6.0 (2.5)2 7.0 (3.0)2 2.9 (0.4) Non-Hispanic black 30.6 (6.3) 3.3 (1.0)1 30.1 (10.2)1 3.0 (1.1)1 3 4.0 (2.0)2 3 2.9 (0.9)1 All Hispanic 3 3.1 (1.2)1 3 3.3 (1.6)2 3 3 3 1.7 (0.7)1 Mexican American 3 3.5 (1.2)1 3 3.3 (1.5)2 3 3 3 1.7 (0.5) Previous diagnosis of stroke is based on self-report. Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% 2 Relative standard error >40%–50% 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Health and Nutrition Examination Surveys 1999–2004

FOOT ULCERS

INTRODUCTION TABLE 20.18. Percent of Adults Age ≥40 Years With Foot Lesions, Overall and by Diabetes Diabetic foot ulcer (DFU) is defined as a Status, U.S., 1999–2004 chronic full thickness defect distal PERCENT (STANDARD ERROR) to the malleoli. It occurs frequently as a of diabetes, with an Overall All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes 2 1 3 estimated lifetime risk of 12%–25% 0.81 (0.20) 0.77 (0.33) 0.73 (0.28) 0.81 (0.23) Foot lesions are defined by the presence of bandages, , ulcers, abrasions, lacerations, and sutures. (63,64,65,66). Reported DFU annual inci- Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose dence ranges from 1.6% (67) to 7.2% (68) ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. for first DFU development and from 7.8% A1c, glycosylated hemoglobin. 1 Relative standard error >30%–40% (69) to 48.0% (70) for DFU recurrence. 2 Relative standard error >40%–50% The reported prevalence of current or past 3 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. history of DFU ranges between 10.4% (71) SOURCE: National Health and Nutrition Examination Surveys 1999–2004 and 57.9% (72). DFU annual incidence from 2006 to 2008 in Medicare benefi- FIGURE 20.14. Percent of Outpatient Visits to a Physician Pertaining to Foot Ulcers Disease, ciaries with diabetes was between 6% and by Diabetes Status, U.S., 2002–2009 approximately 13% in those with diabetes Overall Diabetes No diabetes and PAD (73). According to a new analysis 1 of NHANES 1999–2004 data, 0.77% of all patients with diabetes (including those 0.8 undiagnosed) presented with active foot 0.6 lesions on . For this survey, foot lesions were defined as pres- 0.4 Foot ulcers (%) ence of any of the following: bandages, 0.2 blisters, ulcers, abrasions, lacerations, and sutures. Moreover, those without 0 Diabetes Status diabetes compared to those with diabetes had a higher frequency of foot lesions Foot ulcers are defined based on ICD-9 codes 040.0, 440.24, 785.4, 440.23, and 707.1. Diabetes is defined based on exam (Table 20.18). As the NHANES on ICD-9 codes 250, 357.2, 362.0, 366.41, 648.0, and 775.1. Error bars represent 95% confidence intervals. ICD-9, International Classification of Diseases, Ninth Revision. examiners used a very broad definition All relative standard errors ≤30% of foot lesions that was not specific for SOURCE: National Ambulatory Medical Care Surveys 2002–2009

20–16 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes foot ulcer, the meaning of this finding FIGURE 20.15. Percent of Hospital Discharges Listing Foot Ulcers Disease, by Diabetes is uncertain, but most likely the lesions Status, U.S., 2002–2009 detected were mainly not those that would be typically classified as a foot ulcer Overall Diabetes No diabetes 5 related to diabetes. Conversely, based on International Classification of Diseases, 4 Ninth Revision (ICD-9) codes, new analysis 3 of the National Ambulatory Medical Care Surveys 2002–2009 that captured outpa- 2 tient visits to a physician pertaining to Foot ulcers (%) foot ulcers found a nearly ninefold greater 1 frequency of such visits among persons 0 with compared to those without diabetes Diabetes Status

(Figure 20.14). A similar higher frequency Foot ulcers are defined based on ICD-9 codes 040.0, 440.24, 785.4, 440.23, and 707.1. Diabetes is defined based of hospital discharges listing foot ulcer on ICD-9 codes 250, 357.2, 362.0, 366.41, 648.0, and 775.1. Error bars represent 95% confidence intervals. Confidence intervals were most likely underestimated because the National Hospital Discharge Survey sampling in those with diabetes was seen in the variables were not available, and consequently, it was not possible to take into account the complex sampling design. National Hospital Discharge Surveys from See Table 20.19 for further details. ICD-9, International Classification of Diseases, Ninth Revision. 2002–2009 (Figure 20.15, Table 20.19). All relative standard errors ≤30% Persons with diabetes age 45–64 years SOURCE: National Hospital Discharge Surveys 2002–2009 compared to other age categories, women, and American Indian/Alaska Native ethnic TABLE 20.19. Percent of Hospital Discharges Listing Foot Ulcers, Overall and by Age, Sex, groups had the highest frequencies of Race, and Diabetes Status, U.S., 2002–2009 hospital discharges listing foot ulcer. PERCENT (STANDARD ERROR)

PATHOPHYSIOLOGY Characteristics Overall Diabetes No Diabetes AND RISK FACTORS Overall 1.2 (0.01) 4.2 (0.06) 0.7 (0.01) DFU scarcely occurs due to a single cause Age (years) <45 0.2 (0.01) 3.1 (0.14) 0.1 (0.01) (74,75). Instead, a number of factors 45–64 1.7 (0.03) 4.9 (0.10) 0.9 (0.02) contribute to its development and mainte- 65–74 2.0 (0.04) 4.3 (0.12) 1.2 (0.04) nance. Chronic hyperglycemia progresses ≥75 2.1 (0.03) 3.8 (0.10) 1.7 (0.03) to diabetic (DPN) Sex Men 0.9 (0.01) 3.2 (0.07) 0.6 (0.01) and/or , which in the pres- Women 1.5 (0.02) 5.4 (0.10) 0.9 (0.02) ence of trauma may result in DFU and, in Race advanced cases, lower extremity amputa- White 1.1 (0.02) 4.1 (0.08) 0.7 (0.01) tions (LEA) (64,66,74,76,77,78,79,80). Black 1.5 (0.03) 4.5 (0.14) 0.9 (0.03) AIAN 1.3 (0.19) 4.9 (0.81) 0.6 (0.16) Asian 0.5 (0.06) 2.2 (0.35) 0.3 (0.05) Neuropathy Foot ulcers are defined based on ICD-9 codes 040.0, 440.24, 785.4, 440.23, and 707.1. Diabetes is defined based Chronic hyperglycemia causes changes on ICD-9 codes 250, 357.2, 362.0, 366.41, 648.0, and 775.1. Standard errors were most likely underestimated in function, mainly because the National Hospital Discharge Survey sampling variables were not available, and consequently, it was not possible to take into account the complex sampling design. AIAN, American Indian/Alaska Native; ICD-9, through of the endoneural International Classification of Diseases, Ninth Revision. microvascular circulation (79), damaging SOURCE: National Hospital Discharge Surveys 2002–2009 the nerves (especially those with smaller diameter and less myelination) (65), thus foot muscles, foot deformities, and peripheral motor nerve and cause anterior affecting somatic and autonomic fibers metatarsal verticalization, which create crural muscle atrophy, producing ankle (63,65,79,81,82). The majority of DFUs areas on the foot with elevated pressure equinus. This biomechanical alteration have as a primary cause the presence of peaks (63,65,82). At the same time, the causes an increase in the forefoot pres- DPN (66,79,83,84), which is considered a metatarsal pad is dislocated, reducing sure and shearing forces (65,66,81,82) major factor for their occurrence (74,80). its natural function (66,79,81,82) of and was identified as a precipitant factor Thus, a slower motor nerve conduction dissipating the weight-bearing forces in for the development, recurrence, and velocity has been associated with the all directions (82). This process leads recalcitrance of DFU (66). Changes presence of DFU (67,85,86,87). to a triangular forefoot that is difficult in lower extremity reflexes have been to adapt to regular shoes (79). These associated with DFU in several studies Motor Neuropathy. Motor neuropathy changes increase the risk of dorsal and (88,89,90,91,92,93); however, no studies may lead to paresis, ataxic gait, weak- plantar DFU (81). Less commonly, motor have assessed the association of changes ness and atrophy of the small intrinsic neuropathy can also affect a single major in these reflexes with muscle wasting.

20–17 DIABETES IN AMERICA, 3rd Edition

Autonomic Neuropathy. Autonomic a stocking pattern (81), and diminishes alterations leads to tissue anoxia, cell neuropathy leads to decreased sweating, pain and temperature perception death (125), and DFU (65,75). Noninvasive dry skin, and callus formation in areas (63,81), affecting the protective response testing is crucial, because usual signs are of higher pressure, since the autonomic to potential causes of foot trauma. less frequent in persons with diabetes due nervous system regulates perspiration, Neuropathy symptoms, such as numb- to DPN and more distal arterial skin temperature, and arteriovenous ness, pain, and/or tingling, are associated localization (107,125,126). In persons with shunting. Measurement of autonomic with DFU (89,102,103,104). In fact, all diabetes, complications in the small and function is generally performed by patients with an active or recently healed large vessels frequently do not advance at assessment of cardiovascular reflexes DFU have at least one of these symptoms an equal pace, so one may easily observe that may not reflect function in the (102). toes with ischemic signs caused by small extremities. An indicator test for sudo- vessel alterations, while foot pulses motor function has been associated with Several instruments have been developed may remain intact (125). An association clinical severity of DPN (94) and active to detect loss of protective sensation. between the foot palpable number or recently healed DFU (95), as well as The most widely used are the Semmes- and DFU was reported in several studies excellent reproducibility (94) and high Weinstein monofilament (SWM) and the (88,90,91,104,111). One of two studies accuracy for small-fiber impairment biothesiometer, which were associated reported an association of ABI value detection. Autonomic neuropathy also with DFU in 21 (69,86,88,89,90,91,​92,​ with the development of DFU (89,91), leads to vasodilatation of the dorsal foot 93,​96,​97,102,105,106,107,108,​109,​110,​ whereas an association of ABI value with (63,65,66,74,76,79,81,82,83), and 111,​112,​113,114) and 19 (67,68,72,​84,​85,​ DFU recurrence has not been observed arteriovenous shunts increase pressure 91,​93,102,104,105,107,​111,​114,​115,​116,​ (100,127). Only one study has assessed and arterial flow and may thereby lead 117,118,119,120) studies, respectively. the hallux-brachial index, observing to peripheral and impaired For the SWM, no collection procedure that values ≤0.7 were associated with microvascular response to damage (66). (sites and number of applications) higher rates of active or recently healed Baseline edema was associated with has been widely implemented, and DFU (93). Transcutaneous oxygen pres- DFU development in two studies (89,96) therefore, several studies have been sure is widely used for DFU prognostic conducted by the same group, but not conducted to identify the best cutoff assessment. However, research is scarce in another (97). Furthermore, according (102,107,112,119). The tuning fork is a regarding DFU prediction by this measure to the neurovascular theory, such blood cheaper alternative to the biothesiometer, (89,100,127). flow alteration can lead to bone reab- since it was consistently associated with sorption and weakening and consequent DFU in five studies (88,89,90,92,119). Elevated Pressure Mechanisms Charcot neuroarthropathy. This foot Several methods for thermal sensitivity and Measurement deformity occurs in up to 13% of persons evaluation have been studied (warm Weight, Height, Waist Circumference, with diabetes and DPN (98) and was and cool rods, thermaesthesiometer, and BMI. Weight, height, waist circum- described as being associated with DFU SensortekTM, and other thermal testers), ference, and BMI are related to high foot occurrence (89,99), but not with its recur- and all have been reported to be pressure and macrovascular complica- rence (70,100). With the denervation of associated with DFU (84,85,88,109). tions. In addition, may result in dermal structures, skin loses its integrity, Several scores have been created poor ability to see the feet, which impairs mainly through cracks and fissures that for DPN screening. The Neuropathy foot self-care (126). However, insuffi- facilitate microbial invasion and Symptom Score (88,103,114,117,121) cient evidence is available to support (63,66,80,81,82). Conversely, tinea pedis and the Neuropathy Disability Score the idea that higher weight (67,89,96), presence may be a clinical marker for (88,106,111,114,117,119,122) have waist circumference (71,93), or BMI intact autonomic function due to the need been associated with DFU. The Michigan (85,92,114,128) is related to an increased for a moist medium for fungal growth. Neuropathy Screening Instrument has risk of DFU. On the other hand, height Tinea pedis was associated with a reduced high accuracy for DPN detection (123,124) was associated with DFU in four studies risk for DFU development in two studies and reproducibility (123); however, only (71,89,93,99), which may be due to the (96,97). one study has reported its association observation that longer nerve axons are with DFU development (68). more prone to DPN (71). Sensory Neuropathy. Motor and auto- nomic neuropathy would have a smaller Peripheral Arterial Disease Foot Deformity and Callus. Several impact if it were not for the simulta- PAD rarely leads directly to DFU but foot deformities, such as abnormal neous presence of sensory neuropathy is believed to be a protagonist in the foot (89,96,97,108), rigid toe deformity (66,75,81,82,101), which is responsible pathway to DFU (66,79). PAD diminishes (88,89,108), hallux limitus or rigidus for loss of protective sensation in the oxygen levels in tissues, decreasing (89,108), hallux abductus valgus (108), diabetic foot (79). Sensory neuropathy their resilience (82), which together with subtarsal (108,111) and first metatar- progresses from distal to proximal, in trauma and/or sensory and motor nerve so-phalangeal joints (89,111) mobility

20–18 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes limitation, have been consistently asso- treatment associated with an increased studies (88,104,110,135). Conversely, ciated with DFU development. Evidence risk for DFU (89,90,96,99,103). This one study presents results indicating regarding DFU recurrence and foot defor- association may reflect greater diabetes that nephropathy can be a potential mities is scarce. Hyperkeratotic areas severity in persons treated with insulin confounding variable (144). No studies (callus) are a natural reaction to pressure compared to lifestyle or oral medication. have shown nephropathy to be related to or friction, but they create even more There is no evidence for an association DFU recurrence (100,117,127,138,145). pressure to the subcutaneous tissues, between type of diabetes and the risk Onychomycosis was linked to higher and hemorrhaging into the callosity is a of DFU development. Longer diabetes risk of DFU development in two studies common clinical finding 125( ,129), espe- duration is associated with a higher risk (96,97); however, the use of therapeutic cially in patients with DPN (125). Presence of DPN (133) and/or PAD, and greater nail lacquer did not reduce the risk in a of callus at baseline was associated with diabetes duration also increases the risk randomized controlled trial (RCT) (146). DFU in two studies (99,129), but not in of DFU (71,87,88,89,90,91,95,96,99,​ Previous foot complications (DFU or LEA) another (96); yet, the number of areas 103,104,​105,110,111,118,120,128,​130,​ were consistently related with a new DFU with callus does not seem to influence the 134,​135,​136,137). Poor glucose control occurrence (85,88,89,91,96,97,99,104,​ DFU risk (129). as reflected by higher A1c has been 107,110,111,117,129,134,135,138,141). associated with DFU risk. A majority of The association between previous DFU Pressure Measurements. Subjects with studies showed a statistically significant and new DFU development was observed greater peak plantar pressure values association between A1c value and DFU in a retrospective study even after adjust- consistently have a higher risk of DFU development (85,89,96,97,110,134), but ment for age, sex, visual acuity, physical occurrence (67,104,​105,​111,​116,​129,​130,​ not with DFU recurrence (110,138). impairments, diabetes type and dura- 131). Surprisingly, greater peak plantar tion, PAD complication history, diabetes pressure had no impact in the prediction Physical Impairments complication count, and previous LEA of recurrent DFU (100). Caselli et al. (116) Good visual acuity and physical ability are (147). found that patients with high forefoot peak essential for correct foot self-care (123). pressure and a forefoot/rearfoot ratio >2 Studies analyzing the impact of poor NATIONAL SURVEY RESULTS were more likely to have advanced DPN vision (88,89,96,97), retinopathy (97,134), Self-reported telephone survey data from and a higher risk of DFU development. and laser photocoagulation history (89,96) the Behavioral Risk Factor Surveillance Sauseng et al. (87) showed that maximum reported positive associations between System (BRFSS) were used in new anal- plantar pressure, plantar loading over these variables and risk for DFU. Only one yses for Diabetes in America to assess time, and relative contact time in the first study reported physical impairment to be the frequency of “ever having foot lesions metatarsal head were higher in subjects associated with DFU development (97). that took more than 4 weeks to heal” with a neuropathic plantar DFU. Almost no among persons with diabetes. Such evidence is available on the effect of daily Other Risk Factors lesions have a good likelihood of repre- weight-bearing physical activity on DFU In the great majority of studies, an asso- senting DFU, although no assessment risk. Three studies concluded that less ciation between sex and DFU occurred, has been done to confirm this. One must than average daily activity was associated and men were consistently at higher risk keep in mind that other chronic lower with a higher risk for DFU (69,118,132). for DFU (84,88,95,97,104,105,111,130,​ limb often occur in persons These results are in agreement with the 138,139). Regarding race/ethnicity, several with diabetes, such as venous leg ulcers, “physical stress theory” proposed by studies concluded that whites had higher pressure wounds, and infectious and Maluf and Mueller, which states that a rates of DFU compared with blacks and/or malignant dermatologic . gradual increase in physical stimula- Asians (105,128,135,140,141). No study The overall frequency of self-reported tion leads to plantar protective tissue detected an association between smoking skin lesions did not vary much over hypertrophy, preventing skin breakdown habits (96,97,100,110,113), any of the the BRFSS 2000–2007 cycles (Table (69,118). Conversely, one study (132) items of the lipid profile 100( ,113,127,134), 20.20). A number of characteristics were reported that higher activity variability or low educational attainment associated with higher frequency of represented an increased risk. (71,86,92,97,100,104,128,142) with risk self-reported foot lesions, including age of initial or recurrent DFU. Insufficient <65 years, Hispanic or American Indian Diabetes Characteristics evidence is available regarding the impact ethnicity, insulin treatment, diabetes and Glycemic Control of depression, physical inactivity, onset before age 30 years, current Poorer glycemic control is associated consumption habits, or cardiovascular smoker, not having exercised in the past with the development of several diabetic complications on DFU risk. Several month, higher BMI, and using special foot complications (104,115), but no authors (143) affirm that nephropathy equipment for disability (Table 20.20). clinical trial data are available on intensive should be included in diabetic foot glycemic control and DFU risk. Type of classification, because it has been asso- treatment has been examined, with insulin ciated with DFU occurrence in several

20–19 DIABETES IN AMERICA, 3rd Edition

TABLE 20.20. Percent With Foot Lesions Among Adults Age ≥18 Years With Diagnosed Diabetes, Overall and by Age, Sex, Race/Ethnicity, Insulin Use, Age of Diabetes Onset, Smoking, Alcohol Use, Exercise, BMI, and Disability, U.S., 2000–2007

PERCENT (STANDARD ERROR) CHARACTERISTICS 2000 2001 2002 2003 2004 2005 2006 2007 Overall 12.7 (0.7) 11.5 (0.5) 11.6 (0.7) 13.0 (0.6) 11.9 (0.5) 11.0 (0.4) 11.8 (0.4) 11.3 (0.4) Age (years) 18–39 16.1 (1.9) 11.9 (1.5) 13.0 (2.6) 19.0 (2.8) 12.8 (1.5) 11.6 (1.4) 13.5 (1.6) 13.2 (2.2) 40–64 14.4 (1.1) 12.9 (0.8) 12.9 (0.9) 14.5 (0.8) 13.9 (0.8) 12.6 (0.6) 13.8 (0.7) 13.1 (0.6) 65–74 10.5 (1.1) 9.9 (1.1) 9.1 (1.0) 10.5 (0.9) 9.2 (0.8) 9.3 (0.8) 9.2 (0.7) 8.6 (0.6) ≥75 8.4 (1.0) 8.8 (1.0) 10.0 (2.2) 7.5 (0.7) 8.7 (1.0) 7.6 (0.7) 8.1 (0.8) 8.3 (0.7) Sex Men 12.8 (1.0) 11.6 (0.8) 11.5 (0.9) 14.1 (0.8) 12.7 (0.8) 11.1 (0.6) 11.7 (0.7) 12.0 (0.7) Women 12.6 (0.8) 11.4 (0.7) 11.7 (1.0) 11.9 (0.8) 11.0 (0.6) 10.8 (0.5) 11.8 (0.6) 10.6 (0.5) Race/ethnicity Non-Hispanic white 12.7 (0.6) 10.8 (0.6) 11.1 (0.5) 12.0 (0.5) 11.6 (0.5) 11.1 (0.4) 11.3 (0.5) 10.9 (0.4) Non-Hispanic black 9.1 (1.0) 10.0 (1.4) 8.5 (1.2) 12.4 (1.6) 10.4 (1.1) 10.3 (0.9) 10.3 (0.9) 10.1 (1.0) All Hispanic 15.5 (2.7) 14.7 (1.9) 17.9 (3.3) 15.1 (2.0) 13.1 (1.7) 11.0 (1.6) 13.7 (1.6) 13.8 (1.6) American Indian 15.0 (4.6)1 25.1 (6.4) 10.3 (2.5) 21.9 (6.8)1 14.1 (3.0) 8.6 (2.7)1 17.2 (3.8) 12.8 (2.2) Insulin use Yes 18.7 (1.4) 17.1 (1.3) 17.6 (1.7) 20.7 (1.2) 17.4 (0.9) 17.4 (0.9) 18.8 (1.0) 18.4 (0.9) No 10.4 (0.7) 9.5 (0.6) 9.4 (0.6) 10.3 (0.6) 10.1 (0.6) 8.8 (0.4) 9.4 (0.5) 8.7 (0.4) Age of diabetes onset <30 years 21.1 (2.0) 17.6 (2.0) 17.4 (2.2) 21.9 (2.5) 19.5 (1.6) 19.8 (1.7) 19.7 (1.9) 19.5 (2.2) ≥30 years 11.7 (0.7) 10.6 (0.6) 10.8 (0.7) 11.9 (0.6) 11.0 (0.5) 10.0 (0.4) 10.9 (0.5) 10.4 (0.4) Smoking status Never smokers 10.6 (0.7) 9.9 (0.8) 11.1 (1.1) 11.5 (0.8) 11.4 (0.7) 10.4 (0.6) 11.1 (0.7) 10.8 (0.7) Former smokers 12.7 (1.0) 12.0 (0.9) 11.2 (1.1) 12.1 (0.8) 10.6 (0.7) 10.1 (0.6) 11.3 (0.7) 10.2 (0.5) Current smokers 18.4 (2.4) 14.9 (1.4) 13.5 (1.2) 18.6 (1.7) 16.1 (1.4) 14.5 (1.0) 15.1 (1.0) 15.5 (1.0) Alcohol use <1 drink/day 12.7 (1.2) 11.4 (0.6) 11.8 (0.7) 13.0 (0.6) 11.8 (0.5) 10.9 (0.4) 11.9 (0.5) 11.4 (0.4) ≥1 drink/day 8.4 (3.0)1 9.7 (1.8) 7.6 (1.6) 11.1 (2.5) 14.2 (3.5) 12.5 (2.7) 12.6 (1.9) 10.0 (1.5) Exercise in past month Yes 10.6 (0.8) 9.3 (0.7) 9.7 (0.9) 10.7 (0.6) 10.4 (0.6) 8.7 (0.4) 9.4 (0.5) 9.1 (0.5) No 16.2 (1.2) 14.7 (0.9) 14.6 (1.0) 16.9 (1.1) 14.2 (0.8) 14.1 (0.7) 15.5 (0.9) 14.8 (0.7) BMI (kg/m2) <25 12.7 (1.6) 12.8 (1.5) 10.4 (1.0) 13.8 (1.9) 10.6 (1.0) 11.2 (0.9) 10.6 (1.2) 12.1 (1.4) 25–29.9 10.3 (0.9) 9.9 (0.8) 9.6 (0.9) 12.8 (0.9) 10.5 (0.9) 9.5 (0.6) 11.9 (0.9) 9.9 (0.7) 30–39.9 13.9 (1.3) 11.2 (0.9) 12.0 (1.1) 12.1 (0.8) 12.8 (0.8) 11.1 (0.6) 11.5 (0.6) 11.2 (0.6) ≥40 21.2 (2.8) 15.3 (2.1) 15.7 (2.7) 17.0 (1.7) 15.4 (1.7) 14.4 (1.3) 15.3 (1.3) 14.7 (1.2) Use special equipment for disability* Yes - - - 23.6 (1.5) 22.4 (1.3) 20.7 (1.0) 21.7 (1.3) 20.7 (1.0) No - - - 10.5 (0.6) 9.5 (0.5) 8.5 (0.4) 8.9 (0.4) 8.7 (0.5) Foot lesions were determined based on self-report of ever having lesions that took more than 4 weeks to heal. BMI is based on self-reported height and weight. Diagnosed diabetes is based on self-report. BMI, . * Use of special equipment for disability was only asked in 2003–2007. 1 Relative standard error >30%–40% SOURCE: Behavioral Risk Factor Surveillance System 2000–2007

A NHANES 1999–2004 survey question PREVENTION STRATEGIES DFU (80). This low assessment rate may be on global assessment of health status Guidelines from 2004 advocate that partly explained by the lack of knowledge of was analyzed for Diabetes in America. education should be provided to all diabetic the most important items to assess during Participants with diagnosed diabetes patients on foot care and that they should the screening evaluation (104). In addition, and foot lesions had a higher frequency have their risk status assessed at least evidence regarding the impact of podiatric of fair/poor self-reported health (Table annually (76,148). This recommendation care (91,135,149) and diabetic foot educa- 20.21) than those without foot lesions. is frequently neglected (104,125,142). In tional programs on DFU risk is insufficient Data were too sparse to permit strati- a U.K. study, <20% of diabetic patients (88,150,151,152). Additional data on the fication by age, sex, and race/ethnicity. had their feet examined by a health care utilization of podiatry care in outpatient Similar findings were seen from this professional, and the foot exam annual rate settings in the United States are provided survey for all persons with diabetes with ranged from 30% to 50% in the physician’s in Chapter 39 Medication Use and Self- regard to number of days in the past 30 office 125( ). A complete assessment Care Practices in Persons With Diabetes, days that physical health was not good was performed in only about 10% of the Chapter 40 Health Care Utilization and (Table 20.22). diabetic population in outpatient clinics and Costs of Diabetes, and Chapter 41 Quality in 14% of those admitted to hospital due to of Care in People With Diabetes.

20–20 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

Diabetic Foot Risk Classifications TABLE 20.21. Percent of Adults Age ≥40 Years With Diagnosed Diabetes With Fair or Poor Several classification systems have been Self-Reported Health, by Foot Lesion Status, U.S., 1999–2004 proposed for stratifying patients with PERCENT (STANDARD ERROR) diabetes by risk of DFU development Foot Lesions No Foot Lesions (143). The most widely known (145) are 1 the (1) American Diabetes Association, 53.5 (20.6) 42.4 (2.3) Foot lesions are defined as bandages, blisters, ulcers, abrasions, lacerations, and sutures. Diagnosed diabetes is based (2) International Working Group on on self-report. Diabetic Foot (IWGDF), (3) Scottish 1 Relative standard error >30%–40% Intercollegiate Guidelines Network (SIGN), SOURCE: National Health and Nutrition Examination Surveys 1999–2004 and (4) University of Texas systems, and (5) the Seattle risk score. Despite differ- TABLE 20.22. Mean Days Physical Health Was Not Good in the Past 30 Days Among Adults ences in the number of risk strata and Age ≥40 Years With Diabetes, by Foot Lesion Status, U.S., 1999–2004 number and types of variables included in MEAN (STANDARD ERROR) each system, the majority of the systems Foot Lesions No Foot Lesions had identical core variables: DPN, PAD, 1 foot deformity, and previous diabetic 16.1 (6.9) 5.4 (0.5) Mean days physical health was not good is based on self-report. Foot lesions are defined as bandages, blisters, foot complications (DFU and LEA). A ulcers, abrasions, lacerations, and sutures. Diabetes includes both diagnosed and undiagnosed diabetes. Diagnosed systematic review concluded that the diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated overall evidence quality around these hemoglobin. systems is low, because little validation 1 Relative standard error >40%–50% has been conducted (145). A retrospective SOURCE: National Health and Nutrition Examination Surveys 1999–2004 cohort study of 270 participants with a 1-year follow-up, designed to validate classification proposed by Abbott et al. therefore be of use in identifying persons all of the systems simultaneously, found (68), increased the risk of DFU devel- at risk for this outcome who might benefit no differences in accuracy (153). Further opment in two studies (88,97). The use from preventive measures (164). prospective research to assess the of therapeutic footwear was related described systems’ predictive accuracy to a decrease in DFU development RESEARCH ON CLINICAL COURSE and evaluate new pertinent variables is (89,157) and recurrence (136,154) in Foot Ulcer Healing needed. some studies, but not in all (70,158). DFU is the most frequent cause of LEA All studies assessing the impact of ther- in persons with diabetes (101,165). Each Foot Self-Care Behaviors apeutic shoes compliance verified that DFU requires 15–20 weeks on average and Inspection greater compliance with wearing the to heal, and until re-epithelialization has Evidence of foot self-care impact on DFU recommended footwear was associated occurred, the risk for infection is substan- risk is scarce. One study (97) reported that with a lower rate of DFU occurrence tially increased (165). The time range for poor nail care at baseline was not asso- (141,157,159). healing, though, is skewed. The Eurodiale ciated with DFU development. Irregular study reported that after 1 year, 12% of or insufficient moisturizing represented a Home Temperature Monitoring DFUs were still under treatment (166). higher risk for DFU development in one DFU is accompanied by an inflammatory study (113), but not in another (97). No response, one manifestation of which Classification, Staging, and Treatment association was found between foot self- is a cutaneous temperature increase. Location, Depth, Area, and Duration. care habits, such as washing, sock use, or Foot temperature assessment has been The most frequent location of DFU is the soaking, and the risk of DFU development examined as a potential tool for predicting pulp of the hallux and beneath the first in one study (113). However, patients with complications and leading patients to seek metatarsal (167). DFU located on the toes higher foot care scores had a lower risk medical care (139,160,161,162,163). The present a better prognosis in comparison of DFU recurrence in another study (93). use of self-administered infrared tempera- with those located on other areas of Foot care practices, including self-care ture sensors was related to a significant the foot (168,169,170). A greater depth and exams by a health care provider, are reduction in the risk of DFU development (166,168,171,172,173,174,175), cross-sec- also described in Chapter 39. (161,162) and recurrence (163). However, tional area (166,171,173,175,176,177,178), this tool as a baseline one-time measure- duration at first assessment 166( ,169,173, Footwear ment failed to accurately predict DFU 176,177), and multiple DFU (169,173,176) In several studies, footwear was the development in a 2-year prospective are associated with longer time to heal most frequent precipitating factor for cohort study (160). Nevertheless, a 2013 and poorer prognosis. Initial healing prog- foot ulceration (97,154,155) and half of systematic review and meta-analysis ress with a reduction in area at 1–2 weeks all diabetic amputations (156). The use concluded that this instrument is effective has been associated with greater chance of high-risk footwear, according to a in predicting DFU occurrence and may of complete healing (179).

20–21 DIABETES IN AMERICA, 3rd Edition

Neuropathic, Neuro-Ischemic, and studies validating, refining, and comparing favoring the total contact cast (TCC), Ischemic DFU. Most DFUs can be clas- the existing systems are still needed. The which can reduce pressure at the DFU site sified as neuropathic, neuro-ischemic, or most frequently included and validated by 84%–92%. This technique has consis- ischemic (101), which depends upon the variables were PAD, DFU depth, and infec- tently proved to be the most effective diagnosis of DPN and/or PAD. Therefore, tion (183). regarding healing rates and time to heal both DPN (170,171,172,174,180) and PAD (both in observational studies and RCTs) (66,101,166,168,170,171,172,175,178,​ Other Methods to Predict Outcome. in comparison to other offloading modal- 180,​181,182), diagnosed by absent pulses, Several other factors have a significant ities (191,192) and with other therapeutic ABI <0.7, and/or a transcutaneous partial impact on DFU healing. In the Eurodiale procedures, such as becaplermin, bioen- pressure of oxygen (TcPO2) <40 mmHg, study, end-stage renal disease was associ- gineered tissue, or electrical stimulation are associated with a poorer prognosis. ated with a higher rate of DFU nonhealing (186). A Cochrane review provided further While about 55%–60% of DFU are purely (166). Older (110,166,169,172,177) and support for the use of nonremovable, neuropathic, 35%–45% are caused by male patients presented lower healing pressure relieving casts in the treatment neuropathic and ischemic factors (66,74). rates in several studies (166,173,177). of DFU, finding benefit in healing from In one study, the healing rate of patients Subjects with previous DFU or LEA this intervention (193). Due to time and a with PAD, diagnosed through absence history have lower healing rates (168,180). high level of expertise required, TCC is not of pulses, was comparable to the rate in Chronic hyperglycemia is linked to widely adopted by the medical community patients with DPN and significantly supe- compromised cellular matrix reorganiza- (191). On the other hand, the use of thera- rior to those with both complications (172). tion (186) and function peutic shoes to promote healing still lacks (79), phenomena that would be expected sufficient evidence to prove its effective- Prognostic Systems. A systematic review to impede healing delay and ness (191). identifying the available DFU scoring increase the risk of LEA (66). However, systems concluded that there are a great no RCT has been conducted to deter- A meta-analysis of four RCTs concluded variety of prognostic stratification systems, mine whether intensive glycemic control that the group treated with becaplermin but only a few were validated (183,184). improves healing of DFU (79,187). showed higher healing rates compared The systems considered were: (1) Curative with placebo gel. However, becaplermin is Health Services (CHS); (2) Depth of the Treatment Strategies. Several thera- expensive, and therefore, its widespread Ulcer, Extent of bacterial colonization, peutic technologies have been created use is limited (194,195). Concerning Phase of ulcer and Association aetiology to improve DFU treatment and its prog- skin equivalents, two studies, one using (DEPA); (3) Diabetic Ulcer Severity Score nosis. Results have been discouraging, Dermagraft and another Apligraf, demon- (DUSS); (4) Infectious Diseases Society of with little benefit shown over and above strated that they are safe and effective in America–International Working Group on standard wound care with appropriate the treatment of DFU. Hyperbaric oxygen Diabetic Foot (IDSA-IWGDF); (5) Levine and pressure offloading therapy, which is available in few centers, and O’Neal; (6) Lipsky et al.; (7) Meggit- (188). Debridement should be adequately is potentially effective but very expen- Wagner; (8) Margolis et al.; (9) Perfusion, conducted before any healing technology sive (196). This intervention has shown Extent, Depth/tissue loss, Infection, application (63). In fact, an RCT of beca- some promise in the treatment of DFU, Sensation (PEDIS); (10) Size (Area, Depth), plermin reported that those patients with as a meta-analysis only including RCTs, Sepsis, Arteriopathy, Denervation [S(AD) more frequent debridement presented concluded that subjects undergoing this SAD] system; (11) Saint Elian Wound higher healing rates (189). A systematic therapy were at a reduced risk of major Score System (SEWSS); (12) Scottish review investigating the effect of surgical but not minor LEA (197,198,199). Intercollegiate Guidelines Network debridement on DFU healing identified (SIGN); (13) Site, Ischemia, Neuropathy, five RCTs 190( ) of debridement that Bacterial infection, and Depth (SINBAD); demonstrated better outcomes with more (14) Texas University Classification (TUC); frequent application of this intervention. and (15) Van-Acker/Peter. The SIGN Due to the low number of included studies classification is the only one that was and methodological disparities, the validated, in 2006, for DFU development authors concluded that further research and, in 2007, also for LEA occurrence is needed. prediction (172,185). The most frequently validated systems for DFU healing were Pressure offloading is the other corner- the Meggit-Wagner (n=9), S(AD)SAD stone of appropriate wound care. Unless (n=5), and TUC (n=5), showing lower rates repetitive pressure and shear forces of DFU healing as the systems’ grade are diminished, will be and/or stage increased (183). Accuracy impaired (189). Multiple offloading strat- measures varied greatly, and further egies are available, with published data

20–22 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

LOWER EXTREMITY AMPUTATION

INTRODUCTION FIGURE 20.16. Prevalence of a Primary or Secondary Diagnostic Code for Lower Extremity LEA frequently complicates the clinical Amputation, Medicare Beneficiaries, U.S., 2008 course of diabetes and is often associated with other diabetes complications. In 1997, diabetes accounted for more than half of all nontraumatic LEAs in the United States (200). The magnitude of increase in risk of LEA associated with diabetes was estimated at 7.19 (95% CI 4.61–11.22) among 14,407 subjects in the NHANES Epidemiologic Follow-up Study who were observed prospectively between 1971 and 1992 (201). A similar magnitude eightfold increase in risk of LEA was reported from a population-based cohort study in Sweden (202). Depending on the reason for amputation and the vascular status of the patient, the level may involve toe, partial foot, ankle, below the knee, above the knee, hip dysarticulation, or Prevalence of a primary or secondary diagnostic code for lower extremity amputation among Medicare beneficiaries hemipelvectomy. A national study of U.S. with diabetes continuously enrolled in Parts A and B Fee-for-Service plans for at least 12 months by Dartmouth Atlas of Health Care Hospital Referral Regions (HRR) (www.dartmouthatlas.org). veterans with diabetes from 1998 found SOURCE: Reference 73 that toe amputation was performed most frequently, followed by below the knee FIGURE 20.17. Prevalence of Diabetes Based on Claims Data in a 12-Month Continuous amputation (203). Medicare Enrollment Period, U.S., 2008

The frequency of LEA among persons with diabetes in the United States declined from the 1990s to the 2000s. The Centers for Disease Control and Prevention noted a fall in hospitalizations for LEA per 1,000 persons with diabetes age ≥40 years from 11.2 in 1996 to 3.9 in 2008 based on data from the National Hospital Discharge Survey to capture amputation hospitaliza- tions and National Health Interview Survey data to estimate diabetes prevalence (204). LEA rates were found to fall in all demographic groups considered. Rates were not reported by amputation level, so all levels from toe to hemipelvectomy were included in this analysis. A very Prevalence of two or more claims with ICD-9 codes consistent with diabetes or at least one inpatient claim with similar LEA rate of 4 per 1,000 persons ICD-9 codes consistent with diabetes in the 12-month period of continuous enrollment in Parts A and B Fee-for- Service plans among Medicare beneficiaries by Dartmouth Atlas of Health Care Hospital Referral Regions (HRR) with diabetes in 2008 was estimated from (www.dartmouthatlas.org). ICD-9, International Classification of Diseases, Ninth Revision. U.S. national Medicare Parts A and B data SOURCE: Reference 205 using the same ICD-9 codes to capture amputation, with the exception that the permanent disability as determined by with the higher prevalence of diabetes sample did not include amputations higher the Social Security Administration. In the in the Medicare population in the same than above the knee (73). Persons age ≥65 same Medicare population in 2008, the general regions (Figure 20.17) (205,206). years were predominant in this sample, prevalence of LEA among persons with National variation in amputation inci- but others eligible for Medicare were diabetes was 18 per 1,000 persons (205). dence between 2007 and 2010 was also represented as well, including persons Prevalence of LEA varied nationally, with noted across the United Kingdom (207). with end-stage renal disease, amyotrophic higher rates seen mainly in the South Incidence also varied by race/ethnicity, lateral sclerosis, or who have a severe and and Southwest (Figure 20.16), consistent with lower incidence noted in both Asians

20–23 DIABETES IN AMERICA, 3rd Edition and blacks (207). Amputation rates based TABLE 20.23. Prevalence of Lower Extremity Amputations Determined by Physical Examination on the Medicare Diabetes Analytics File Among Adults Age ≥40 Years, Overall and by Diabetes Status, U.S., 1999–2004 in the United States and a Department of PERCENT (STANDARD ERROR) Veterans Affairs study, though, found the opposite results with regard to blacks who AMPUTATIONS Overall All Diabetes No Diabetes had a higher LEA frequency (203,208). Overall 0.18 (0.07)1 0.28 (0.11)1 0.17 (0.07)2 n 13 9 4 Type of diabetes and amputation level are n toe 4 3 1 n partial foot 0 0 0 typically not reported in national surveys n entire foot 9 6 3 of LEA incidence and prevalence. Data Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose from the NHANES 1999–2004 were ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. examined for Diabetes in America to 1 Relative standard error >30%–40% estimate prevalence of amputation by 2 Relative standard error >40%–50% level as assessed by physical examina- SOURCE: National Health and Nutrition Examination Surveys 1999–2004 tion (Table 20.23). Overall prevalence of amputation was 0.18%, but this estimate TABLE 20.24. Percent of Outpatient Visits to a Physician Pertaining to Nontraumatic is imprecise due to the small number of Amputations, by Diabetes Status, U.S., 2002–2009 participants noted to have had an LEA. PERCENT (STANDARD ERROR) The majority of amputations involved at Overall Diabetes No Diabetes least the entire foot. The exact level at the 1 foot or a more proximal location was not 0.02 (0.004) 0.10 (0.044) 0.02 (0.003) Amputation is defined based on ICD-9 codes V49.7 and V52.1. Diabetes is defined based on ICD-9 codes 250, 357.2, specified. A higher prevalence was seen 362.0, 366.41, 648.0, and 775.1. ICD-9, International Classification of Diseases, Ninth Revision. among persons with diagnosed and undi- 1 Relative standard error >40%–50% agnosed diabetes, but statistical inference SOURCE: National Ambulatory Medical Care Surveys 2002–2009 is not possible in these data due to the large standard errors. According to a new TABLE 20.25. Percent of Hospital Discharges Listing Nontraumatic Amputations, Overall analysis of survey data, the percentage and by Age, Sex, Race, and Diabetes Status, U.S., 2002–2009 of outpatient physician visits related to PERCENT (STANDARD ERROR) amputation were infrequent overall in the National Ambulatory Medical Care CHARACTERISTICS Overall Diabetes No Diabetes Surveys 2002–2009 but five times more Overall 0.19 (0.01) 0.72 (0.03) 0.11 (0.004) likely to occur in persons with compared Age (years) to those without diabetes (Table 20.24). <45 0.05 (0.003) 0.43 (0.05) 0.04 (0.003) 45–64 0.34 (0.01) 0.93 (0.05) 0.18 (0.01) 65–75 0.33 (0.02) 0.79 (0.06) 0.18 (0.02) PATHOPHYSIOLOGY ≥75 0.25 (0.01) 0.55 (0.04) 0.18 (0.01) The main indication for LEA is tissue Sex nonviability due to ischemia, infection, or Men 0.12 (0.01) 0.49 (0.03) 0.07 (0.004) Women 0.29 (0.01) 1.00 (0.05) 0.17 (0.01) injury (209). The effect of diabetes is one Race step removed but acts through pathways White 0.18 (0.01) 0.65 (0.04) 0.11 (0.01) that increase risk of ischemia due to PAD Black 0.32 (0.02) 0.98 (0.07) 0.18 (0.01) and infection following the development AIAN 0.53 (0.12) 1.90 (0.46) 0.25 (0.11)1 1 2 2 of a foot ulcer (210). LEA in diabetes is Asian 0.06 (0.03) Amputations are defined based on ICD-9 codes V49.7 and V52.1. Diabetes is defined based on ICD-9 codes 250, frequently preceded by a nonhealing foot 357.2, 362.0, 366.41, 648.0, and 775.1. Standard errors were most likely underestimated because the National ulcer that may lead to extensive infection Hospital Discharge Survey sampling variables were not available, and consequently, it was not possible to take into account the complex sampling design. AIAN, American Indian/Alaska Native; ICD-9, International Classification of involving soft tissue and bone for which Diseases, Ninth Revision. the only effective treatment is amputa- 1 Relative standard error >40%–50% tion (210). Severe PAD associated with 2 Estimate is too unreliable to present; ≤1 case or relative standard error >50%. SOURCE: National Hospital Discharge Surveys 2002–2009 diabetes may require amputation as treat- ment for a nonhealing ulcer, gangrene, or Risk Factors (73,180,211,212). These risk factors refractory pain (209). History of DFU and LEA risk is similar in persons with type 1 closely parallel those of DFU, as described PAD appear to have independent roles in or type 2 diabetes (211,212). The major elsewhere in this chapter. Some racial/ predicting amputation risk, as a positive reported risk factors for amputation ethnic differences have been demon- DFU history was linked to a higher risk include diabetes severity and duration, strated in risk of LEA. Compared to of LEA even when adjusted for PAD and neuropathy, PAD, advanced age, and European/Caucasians, American Indians number of diabetic complications (147). presence of other diabetic complications are at higher risk and South and East Asian

20–24 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes men and women and Afro-Caribbean toe amputation will carry a higher risk of Mortality. Both prevalent and incident men at lower risk (213,214,215). The inci- reamputation of the ipsilateral limb than a LEA were associated with a high annual dence of LEA per 1,000 persons in 2008 more proximal (nearer the hip) amputation, risk of death in the Medicare popula- assessed in the U.S. Medicare population such as trans-tibial, because the more tion in 2008 of 170 and 206 per 1,000, differed by race/ethnicity as follows: distal (nearer the toe) procedure may respectively, compared to the Medicare white, 4; black, 7; Asian, 2; Hispanic, 5; not be adequate to resect the diseased population without LEA (73). A retro- and American Indian/Alaska Native, 8 area to permit healing. Since more limb spective study conducted in the United (73). Smoking was not related to ampu- remains, the potential for additional Kingdom found a similar 1-year mortality tation risk among persons with diabetes amputation is higher. Prior amputation of 170 per 1,000 among persons with in several cohort studies (180,201,216), was associated with an approximately diabetes who had undergone LEA (224). despite the strong association between threefold increase in risk of subsequent However, the association between higher this habit and PAD (217). amputation in a model that controlled mortality and having experienced a LEA for PAD, neuropathy, diabetes duration, is inconsistent, as a study in Fremantle, Regarding the association of LEA with and treatment with insulin, but risk in Australia, found no difference in the risk of glycemia, a meta-analysis based on relation to level of initial amputation cardiac death between diabetic persons 94,640 subjects from 14 prospective was not examined (180). A systematic with and without LEA followed longitudi- studies estimated a 1.26-fold increase in review of the reamputation rate following nally after adjustment for other risk factors LEA risk in relation to each 1% increase a limb-salvaging ray resection (toe and for CVD and diabetes-related complica- in A1c (211). A study based at a large metatarsal) yielded only five studies that tions (225). Northern California health plan confirmed included a total of 435 patients under- an increase in amputation risk not only going this procedure, among whom 86 with elevated A1c but also with higher (19.8%) required reamputation (218). A serum concentration (216). higher risk of ipsilateral reamputation The DCCT and the UKPDS assessed among patients who had undergone more outcomes in relation to an intensive distal amputation was generally seen in compared to a standard glucose control other investigations (219,220). A high risk strategy. Neither study to date has reported of amputation to the contralateral limb whether this intervention resulted in fewer at 5-year follow-up varied by level of the amputations or foot ulcers. initial amputation, with risk ranging from 18.8% for an initial toe to 53.3% for an National Hospital Discharge Survey data initial below the knee amputation. from 2002–2009 were assessed for Diabetes in America to identify character- Functional Status. Little information istics associated with amputation (Table is available to assess the effects of LEA 20.25). The proportion of discharges for on functional status and quality of life amputation among persons with diabetes among persons with diabetes specifically. was approximately sevenfold higher In general, diabetes is associated with compared to persons without diabetes. functional impairment, as assessed by the The highest proportion of discharges for Medical Outcomes Study Short Form 20 amputation among persons with diabetes (221). Mexican Americans with diabetes occurred in the 45–64-years age range. had an approximately twofold increase in This pattern was not seen among persons the risk of having a significant impairment without diabetes. Similar patterns of sex in a lower body activity of daily living (higher in women) and race (higher in compared to persons without diabetes. black and American Indian/Alaska Native) Perhaps not surprisingly, the presence of differences were seen between persons an LEA increased the risk of any significant with and without diabetes. lower body impairment 2.3-fold among those with diabetes in this population Outcomes of LEA (222). Diabetic LEA was associated with a Reamputation. An initial LEA increases significantly higher Sickness Impact Profile the risk of subsequent amputation to the score among persons with diabetes, but ipsilateral (same) or contralateral (other) this difference was primarily due to poorer limb. Estimates of the degree of this physical dimension scores, as psychosocial risk vary depending on the extent of the dimension scores did not significantly initial surgical procedure. For example, a differ by amputation status (223).

20–25 DIABETES IN AMERICA, 3rd Edition

CONVERSIONS Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions.

LIST OF ABBREVIATIONS A1c ...... glycosylated hemoglobin IL ...... interleukin ABI ...... ankle-brachial index IWGDF ...... International Working Group on Diabetic Foot BMI ...... body mass index LEA ...... lower extremity amputation BRFSS ...... Behavioral Risk Factor Surveillance System NHANES . . . . National Health and Nutrition Examination CI ...... confidence interval Survey CVD ...... cardiovascular disease NO ...... nitric oxide DCCT ...... Diabetes Control and Complications Trial PAD ...... peripheral arterial disease DFU ...... diabetic foot ulcer RCT ...... randomized controlled trial DPN ...... diabetic peripheral neuropathy S(AD)SAD . . . Size (Area, Depth), Sepsis, Arteriopathy, EDIC ...... Epidemiology of Diabetes Interventions and Denervation Complications study SIGN ...... Scottish Intercollegiate Guidelines Network eGFR ...... estimated glomerular filtration rate SWM ...... Semmes-Weinstein monofilament HDL ...... high-density lipoprotein TCC ...... total contact casting ICD-9 ...... International Classification of Diseases, Ninth TUC ...... Texas University Classification Revision UKPDS ...... United Kingdom Prospective Diabetes Study

ACKNOWLEDGMENTS/FUNDING DUALITY OF INTEREST Drs. Boyko and Wheeler were supported by the Veterans Affairs Puget Sound Health Drs. Boyko, Monteiro-Soares, and Care System, Seattle, WA. Drs. Boyko and Wheeler also acknowledge the support of Wheeler reported no conflicts of the Diabetes Research Center at the University of Washington, which is supported interest. by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases (DK017047). Dr. Monteiro-Soares was funded by a grant from Fundação para a Ciência e Tecnologia (FCT), Portugal (SFRH/BD//86201/2012).

REFERENCES

1. American Diabetes Association: Peripheral 6. Halperin JL: Arterial Obstructive Diseases chronic arterial insufficiency of the lower arterial disease in people with diabetes. of the Extremities. Loscalzo CM, Dzau VJ, extremities: a critical review. Circulation Diabetes Care 26:3333–3341, 2003 Eds. Boston, MA, Little Brown, 1992 94:3026–3049, 1996 2. Beckman JA, Creager MA, Libby P: 7. Strandness DE, Jr., Priest RE, Gibbons GE: 11. Rose GA: The diagnosis of ischaemic Diabetes and atherosclerosis: epidemi- Combined clinical and pathologic study of heart pain and intermittent claudication ology, pathophysiology, and management. diabetic and nondiabetic peripheral arte- in field surveys. Bull World Health Organ JAMA 287:2570–2581, 2002 rial disease. Diabetes 13:366–372, 1964 27:645–658, 1962 3. Goldin A, Beckman JA, Schmidt AM, 8. Jude EB, Oyibo SO, Chalmers N, Boulton 12. Belch JJ, Topol EJ, Agnelli G, Bertrand M, Creager MA: Advanced glycation end AJ: Peripheral arterial disease in diabetic Califf RM, Clement DL, Creager MA, Easton products. Sparking the development and nondiabetic patients: a comparison JD, Gavin JR, 3rd, Greenland P, Hankey of diabetic vascular injury. Circulation of severity and outcome. Diabetes Care G, Hanrath P, Hirsch AT, Meyer J, Smith 114:597–605, 2006 24:1433–1437, 2001 SC, Sullivan F, Weber MA; Prevention 4. Forstermann U, Munzel T: Endothelial 9. LoGerfo FW, Coffman JD: Current of Atherothrombotic Disease Network: in vascular disease: concepts. Vascular and microvascular Critical issues in peripheral arterial disease from marvel to menace. Circulation disease of the foot in diabetes. detection and management: a call to action. 113:1708–1714, 2006 Implications for foot care. N Engl J Med Arch Intern Med 163:884–892, 2003 5. Vehkavaara S, Seppala-Lindroos A, 311:1615–1619, 1984 13. Grenon SM, Gagnon J, Hsiang Y: Video in Westerbacka J, Groop PH, Yki-Jarvinen H: 10. Weitz JI, Byrne J, Clagett GP, Farkouh ME, clinical medicine. Ankle-brachial index for In vivo charater- Porter JM, Sackett DL, Strandness DE, assessment of peripheral arterial disease. izes patients with . Jr., Taylor LM: Diagnosis and treatment of N Engl J Med 361:e40, 2009 Diabetes Care 22:2055–2060, 1999

20–26 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

14. Young MJ, Adams JE, Anderson GF, 25. Alexander CM, Landsman PB, Teutsch 33. Kannel WB, McGee D, Gordon T: A Boulton AJ, Cavanaugh PR: Medial arte- SM: Diabetes mellitus, impaired fasting general cardiovascular risk profile: the rial calcification in the feet of diabetic glucose, atherosclerotic risk factors, and Framingham Study. Am J Cardiol 38:46– patients and matched non-diabetic control prevalence of coronary heart disease. Am 51, 1976 subjects. Diabetologia 36:615–621, 1993 J Cardiol 86:897–902, 2000 34. Kannel WB, McGee D, Gordon T: Update 15. Expert Committee on the Diagnosis and 26. Orchard TJ, Dorman JS, Maser RE, Becker on some epidemiologic features of inter- Classification of Diabetes Mellitus: Report DJ, Drash AL, Ellis D, LaPorte RE, Kuler mittent claudication: the Framingham of the expert committee on the diagnosis LH: Prevalence of complications in Study. J Am Geriatr Soc 33:13–18, 1985 and classification of diabetes mellitus. IDDM by sex and duration. Pittsburgh 35. Newman AB, Shemanski L, Manolio Diabetes Care 26(Suppl 1):S5–S20, 2003 Epidemiology of Diabetes Complications TA, Cushman M, Mittelmark M, Polak JF, 16. Potier L, Abi Khalil C, Mohammedi K, Study II. Diabetes 39:1116–1124, 1990 Powe NR, Siscovick D; The Cardiovascular Roussel R: Use and utility of ankle brachial 27. Nathan DM, Lachin J, Cleary P, Orchard Health Study Group: Ankle-arm index as index in patients with diabetes. Eur J Vasc T, Brillon DJ, Backlund JY, O’Leary a predictor of cardiovascular disease and Endovasc Surg 41:110–116, 2011 DH, Genuth S; Diabetes Control and mortality in the Cardiovascular Health 17. Criqui MH, Denenberg JO, Langer RD, Complications Trial; Epidemiology of Study. Arterioscler Thromb Vasc Biol Fronek A: The epidemiology of peripheral Diabetes Interventions and Complications 19:538–545, 1999 arterial disease: importance of identifying Research Group: Intensive diabetes 36. Fabsitz RR, Sidawy AN, Go O, Lee ET, the population at risk. Vasc Med 2:221– therapy and carotid intima-media thick- Welty TK, Devereux RB, Howard BV: 226, 1997 ness in type 1 diabetes mellitus. N Engl J Prevalence of peripheral arterial disease 18. Leng GC, Fowkes FG: The Edinburgh Med 348:2294–2303, 2003 and associated risk factors in American Claudication Questionnaire: an improved 28. Fowkes FG, Rudan D, Rudan I, Aboyans V, Indians: the Strong Heart Study. Am J version of the WHO/Rose Questionnaire Denenberg JO, McDermott MM, Norman Epidemiol 149:330–338, 1999 for use in epidemiological surveys. J Clin PE, Sampson UK, Williams LJ, Mensah 37. Abbott RD, Rodriguez BL, Petrovitch H, Epidemiol 45:1101–1109, 1992 GA, Criqui MH: Comparison of global Yano K, Schatz IJ, Popper JS, Masaki KH, 19. Orchard TJ, Strandness DE, Jr.: estimates of prevalence and risk factors Ross GW, Curb JD: Ankle-brachial blood Assessment of peripheral vascular disease for peripheral artery disease in 2000 and pressure in elderly men and the risk of in diabetes. Report and recommendations 2010: a systematic review and analysis. stroke: the Honolulu Heart Program. J of an international workshop sponsored Lancet 382:1329–1340, 2013 Clin Epidemiol 54:973–978, 2001 by the American Diabetes Association 29. Al-Delaimy WK, Merchant AT, Rimm EB, 38. Uusitupa MI, Niskanen LK, Siitonen O, and the American Heart Association Willett WC, Stampfer MJ, Hu FB: Effect Voutilainen E, Pyorala K: 5-year incidence September 18–20, 1992 New Orleans, of type 2 diabetes and its duration on the of atherosclerotic vascular disease in rela- Louisiana. Circulation 88:819–828, 1993 risk of peripheral arterial disease among tion to general risk factors, insulin level, 20. Bernstein EF, Fronek A: Current status men. Am J Med 116:236–240, 2004 and abnormalities in lipoprotein compo- of noninvasive tests in the diagnosis of 30. Adler AI, Stevens RJ, Neil A, Stratton sition in non-insulin-dependent diabetic peripheral arterial disease. Surg Clin IM, Boulton AJ, Holman RR: UKPDS and nondiabetic subjects. Circulation North Am 62:473–487, 1982 59: Hyperglycemia and other potentially 82:27–36, 1990 21. Hirsch AT, Criqui MH, Treat-Jacobson D, modifiable risk factors for peripheral 39. Tsai AW, Folsom AR, Rosamond WD, Regensteiner JG, Creager MA, Olin JW, vascular disease in type 2 diabetes. Jones DW: Ankle-brachial index and 7-year Krook SH, Hunninghake DB, Comerota Diabetes Care 25:894–899, 2002 ischemic stroke incidence: the ARIC study. AJ, Walsh ME, McDermott MM, Hiatt 31. Hirsch AT, Allison MA, Gomes AS, Corriere Stroke 32:1721–1724, 2001 WR: Peripheral arterial disease detection, MA, Duval S, Ershow AG, Hiatt WR, Karas 40. Aschoff L: Observations concerning the awareness, and treatment in primary care. RH, Lovell MB, McDermott MM, Mendes relationship between cholesterol metab- JAMA 286:1317–1324, 2001 DM, Nussmeier NA, Treat-Jacobson D; olism and vascular disease. Br Med J 22. Nelson KM, Reiber G, Kohler T, Boyko EJ: American Heart Association Council on 2:1131–1134, 1932 Peripheral arterial disease in a multiethnic Peripheral Vascular Disease; Council 41. Fowkes FG, Housley E, Riemersma RA, national sample: the role of conventional on Cardiovascular Nursing; Council on Macintyre CC, Cawood EH, Prescott RJ, risk factors and allostatic load. Eth Dis Cardiovascular Radiology and Intervention; Ruckley CV: Smoking, , glucose toler- 17:669–675, 2007 Council on Cardiovascular Surgery and ance, and blood pressure as risk factors 23. Selvin E, Erlinger TP: Prevalence of Anesthesia; Council on Clinical ; for peripheral atherosclerosis compared and risk factors for peripheral arterial Council on Epidemiology and Prevention: with ischemic heart disease in the disease in the United States: results A call to action: women and peripheral Edinburgh Artery Study. Am J Epidemiol from the National Health and Nutrition artery disease: a scientific statement 135:331–340, 1992 Examination Survey, 1999–2000. from the American Heart Association. 42. Malinow MR, Kang SS, Taylor LM, Circulation 110:738–743, 2004 Circulation 125:1449–1472, 2012 Wong PW, Coull B, Inahara T, Mukerjee 24. Murabito JM, Evans, JC, Nieto K, Larson 32. Newman AB, Siscovick DS, Monolio TA, D, Sexton G, Upson B: Prevalence of MG, Levy D, Wilson PW: Prevalence and Polak J, Fried LP, Borhani NO, Wolfson hyperhomocyst(e)inemia in patients with clinical correlates of peripheral arterial SK: Ankle-arm index as a marker of peripheral occlusive disease. Circulation disease in the Framingham Offspring atherosclerosis in the Cardiovascular 79:1180–1188, 1989 Study. Am Heart J 143:961–965, 2002 Health Study. Cardiovascular Heart Study 43. Robinson K, Arheart K, Refsum H, (CHS) Collaborative Research Group. Brattstrom L, Boers G, Ueland P, Rubba Circulation 88:837–845, 1993 P, Palma-Reis R, Meleady R, Daly L, Witteman J, Graham I: Low circulating folate and vitamin B6 concentrations:

20–27 DIABETES IN AMERICA, 3rd Edition

risk factors for stroke, peripheral vascular long-term prognostic value of the resting Comprehensive foot examination and disease, and coronary artery disease. and postexercise ankle-brachial index. risk assessment: a report of the task European COMAC Group. Circulation Arch Intern Med 166:529–535, 2006 force of the foot care interest group 97:437–443, 1998 54. Wattanakit K, Folsom AR, Criqui MH, of the American Diabetes Association, 44. Lowe GD, Fowkes FG, Dawes J, Donnan Kramer HJ, Cushman M, Shea S, Hirsch with endorsement by the American PT, Lennie SE, Housley SE: Blood AT: Albuminuria and peripheral arterial Association of Clinical Endocrinologists. viscosity, fibrinogen, and activation of disease: results from the Multi-Ethnic Diabetes Care 31:1679–1685, 2008 and leukocytes in peripheral Study of Atherosclerosis (MESA). 65. Fard AS, Esmaelzadeh M, Larijani B: arterial disease and the normal population Atherosclerosis 201:212–216, 2008 Assessment and treatment of diabetic foot in the Edinburgh Artery Study. Circulation 55. Baber U, Mann D, Shimbo D, Woodward ulcer. Int J Clin Pract 61:1931–1938, 2007 87:1915–1920, 1993 M, Olin JW, Muntner P: Combined role of 66. Frykberg RG, Zgonis T, Armstrong 45. Lee AJ, Fowkes FG, Lowe GD, Rumley reduced estimated glomerular filtration DG, Driver VR, Giurini JM, Kravitz SR, A: Fibrin D-dimer, haemostatic factors rate and microalbuminuria on the preva- Landsman AS, Lavery LA, Moore JC, and peripheral arterial disease. Thromb lence of peripheral arterial disease. Am J Schuberth JM, Wukich DK, Andersen C, Haemost 74:828–832, 1995 Cardiol 104:1446–1451, 2009 Vanore JV; American College of Foot and 46. Lee AJ, MacGregor AS, Hau CM, Price JF, 56. Wattanakit K, Folsom AR, Selvin E, Coresh Ankle Surgeons: Diabetic foot disorders. A Rumley A, Lowe GD, Fowkes FG: The role of J, Hirsch AT, Weatherley BD: Kidney func- clinical practice guideline (2006 revision). haematological factors in diabetic peripheral tion and risk of peripheral arterial disease: J Foot Ankle Surg 45(5 Suppl):S1–S66, arterial disease: the Edinburgh Artery results from the Atherosclerosis Risk in 2006 Study. Br J Haematol 105:648–654, Communities (ARIC) Study. J Am Soc 67. Kastenbauer T, Sauseng S, Sokol G, 1999 Nephrol 18:629–636, 2007 Auinger M, Irsigler K: A prospective study 47. Vu JD, Vu JB, Pio J, Malik S, Franklin SS, 57. Nathan DM, Lachin J, Cleary P, Orchard of predictors for foot ulceration in type Chen RS, Wong ND: Impact of C-reactive T, Brillon DJ, Backlund JY, O’Leary 2 diabetes. J Am Podriatr Med Assoc protein on the likelihood of peripheral DH, Genuth S; Diabetes Control and 91:343–350, 2001 arterial disease in United States adults Complications Trial; Epidemiology of 68. Abbott CA, Vileikyte L, Williamson S, with the metabolic syndrome, diabetes Diabetes Interventions and Complications Carrington AL, Boulton AJ: Multicenter mellitus, and preexisting cardiovascular Research Group: Intensive diabetes study of the incidence of and predictive disease. Am J Cardiol 96:655–658, 2005 therapy and carotid intima-media thick- risk factors for diabetic neuropathic ulcer- 48. Wildman RP, Muntner P, Chen J, Sutton- ness in type 1 diabetes mellitus. N Engl J ation. Diabetes Care 21:1071–1075, 1998 Tyrrell K, He J: Relation of inflammation to Med 348:2294–2303, 2003 69. Lemaster JW, Reiber GE, Smith D, peripheral arterial disease in the National 58. Jelnes R, Gaardsting O, Hougaard Jensen Heagerty PJ, Wallace C: Daily weight- Health and Nutrition Examination Survey, K, Baekgaard N, Tonnesen KH, Schroeder bearing activity does not increase the risk 1999–2002. Am J Cardiol 96:1579–1583, T: Fate in intermittent claudication: of diabetic foot ulcers. Med Sci Sports 2005 outcome and risk factors. Br Med J (Clin Exerc 35:1093–1099, 2003 49. McDermott MM, Guralnik JM, Corsi A, Res Ed) 293:1137–1140, 1986 70. Kloos C, Hagen F, Lindloh C, Braun A, Albay M, Macchi C, Bandinelli S, Ferrucci 59. Criqui MH, Langer RD, Fronek A, Leppert K, Muller N, Wolf G, Muller UA: L: Patterns of inflammation associated Feigelson HS, Klauber MR, McCann TJ, Cognitive function is not associated with with peripheral arterial disease: the Browner D: Mortality over a period of 10 recurrent foot ulcers in patients with InCHIANTI study. Am Heart J 150:276– years in patients with peripheral arterial diabetes and neuropathy. Diabetes Care 281, 2005 disease. N Engl J Med 326:381–386, 32:894–896, 2009 50. Navas-Acien A, Selvin E, Sharrett AR, 1992 71. Iversen MM, Midthjell K, Ostbye T, Tell GS, Calderon-Aranda E, Silbergeld E, Guallar E: 60. Howell MA, Colgan MP, Seeger RW, Clipp E, Sloane R, Nortvedt M, Uhlving S, Lead, cadmium, smoking, and increased Ramsey DE, Sumner DS: Relationship of Hanestad BR: History of and factors asso- risk of peripheral arterial disease. severity of lower limb peripheral vascular ciated with diabetic foot ulcers in Norway: Circulation 109:3196–3201, 2004 disease to mortality and morbidity: a the Nord-Trondelag Health Study. Scand J 51. Resnick HE, Lindsay RS, McDermott six-year follow-up study. J Vasc Surg 36:62–68, 2008 MM, Devereux RB, Jones KL, Fabsitz 9:691–696, 1989 72. Lott DJ, Zou D, Mueller MJ: Pressure RR, Howard BV: Relationship of high and 61. Szilagyi DE, Hageman JH, Smith RF, gradient and subsurface shear stress on low ankle brachial index to all-cause and Elliott JP, Brown F, Dietz P: Autogenous the neuropathic forefoot. Clin Biomech cardiovascular mortality. The Strong Heart grafting in femoropopliteal athero- (Bristol, Avon) 23:342–348, 2008 Study. Circulation 109:733–739, 2004 sclerosis: the limits of its effectiveness. 73. Data Points Publication Series [Internet]: 52. O’Hare AM, Vittinghoff E, Hsia J, Shlipak Surgery 86:836–851, 1979 Incidence of diabetic foot ulcer and lower MG: Renal insufficiency and the risk of 62. DeWeese JA, Rob CG: Autogenous venous extremity amputation among Medicare lower extremity peripheral arterial disease: grafts ten years later. Surgery 82:755– beneficiaries, 2006 to 2008: Data Points results from the Heart and / 784, 1977 #2 [article online], 2011. Available from Progestin Replacement Study (HERS). J 63. Brem H, Sheehan P, Rosenberg HJ, http://www.ncbi.nlm.nih.gov/books/ Am Soc Nephrol 15:1046–1051, 2004 Schneider JS, Boulton AJ: Evidence-based NBK65149. 53. Feringa HH, Bax JJ, van Waning VH, protocol for diabetic foot ulcers. Plast 74. Boulton AJ: The diabetic foot: from art to Boersma E, Elhendy A, Schouten O, Reconstr Surg 117:193S–209S, 2006 science. The 18th Camillo Golgi lecture. Tangelder MJ, van Sambeek MH, van 64. Boulton AJ, Armstrong DG, Albert SF, Diabetologia 47:1343–1353, 2004 den Meiracker AH, Poldermans D: The Frykberg RG, Hellman R, Kirkman MS, 75. Apelqvist J, Bakker K, van Houtum Lavery LA, Lemaster JW, Mills JL, Sr., WH, Nabuurs-Franssen MH, Shaper Mueller MJ, Sheehan P, Wukich DK: NC; International Working Group on the

20–28 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

Diabetic Foot: International consensus new diabetic foot ulceration in a commu- 99. Bresater LE, Welin L, Romanus B: Foot and practical guidelines on the manage- nity-based patient cohort. Diabet Med and risk factors for diabetic ment and the prevention of the diabetic 19:377–384, 2002 foot disease in elderly men. Diabetes Res foot. Diabetes Metab Res Rev 16:S84– 89. Boyko EJ, Ahroni JH, Stensel V, Forsberg Clin Pract 32:103–109, 1996 S92, 2000 RC, Davignon DR, Smith DG: A prospec- 100. Peters EJ, Armstrong DG, Lavery LA: Risk 76. Boulton AJM: The diabetic foot. Medicine tive study of risk factors for diabetic foot factors for recurrent diabetic foot ulcers: 34:87–90, 2006 ulcer. The Seattle Diabetic Foot Study. site matters. Diabetes Care 30:2077– 77. Boulton AJ, Armstrong DG, Albert SF, Diabetes Care 22:1036–1042, 1999 2079, 2007 Frykberg RG, Hellman R, Kirkman MS, 90. de Sonnaville JJ, Colly LP, Wijkel D, Heine 101. Apelqvist J, Bakker K, van Houtum Lavery LA, LeMaster JW, Mills JL, Sr., RJ: The prevalence and determinants of WH, Schaper NC; International Working Mueller MJ, Sheehan P, Wukich DK; Task foot ulceration in type II diabetic patients Group on the Diabetic Foot (IWGDF) Force of the Foot Care Interest Group in a primary health care setting. Diabetes Editorial Board: Practical guidelines of the American Diabetes Association: Res Clin Pract 35:149–156, 1997 on the management and prevention Comprehensive foot examination and risk 91. McGill M, Molyneaux L, Yue DK: Which of the diabetic foot: based upon the assessment. Endocr Pract 14:576–583, diabetic patients should receive podiatry International Consensus on the Diabetic 2008 care? An objective analysis. Intern Med J Foot (2007). Diabetes Metab Res Rev 78. Khanolkar MP, Bain SC, Stephens JW: The 35:451–456, 2005 24(Suppl 1):S181–S187, 2008 diabetic foot. QJM 101:685–695, 2008 92. McNeely MJ, Boyko EJ, Ahroni JH, 102. Armstrong DG, Lavery LA, Vela SA, 79. Bowering CK: Diabetic foot ulcers. Stensel VL, Reiber GE, Smith DG, Quebedeaux TL, Fleischli JG: Choosing a Pathophysiology, assessment, and therapy. Pecoraro RF: The independent contri- practical screening instrument to identify Can Fam Physician 47:1007–1016, 2001 butions of and patients at risk for diabetic foot ulceration. 80. Morbach S: Diagnosis, Treatment and vasculopathy in foot ulceration. How great Arch Intern Med 158:289–292, 1998 Prevention of Diabetic Foot Syndrome. are the risks? Diabetes Care 18:216–219, 103. Gulliford MC, Mahabir D: Diabetic foot Heidenheim, Germany, Paul Hartmann 1995 disease and foot care in a Caribbean AG, 2003 93. Porciuncula MV, Rolim LC, Garofolo community. Diabetes Res Clin Pract 81. Laughlin RT, Calhoun JH, Mader JT: The L, Ferreira SR: [Analysis of factors 56:35–40, 2002 diabetic foot. J Am Acad Orthop Surg associated with extremity ulceration in 104. Lavery LA, Armstrong DG, Vela SA, 3:218–225, 1995 diabetic subjects with peripheral neurop- Quebedeaux TL, Fleischli JG: Practical 82. Sumpio BE: Foot ulcers. N Engl J Med athy]. [Article in Portuguese] Arq Bras criteria for screening patients at high risk 343:787–793, 2000 Endocrinol Metabol 51:1134–1142, 2007 for diabetic foot ulceration. Arch Intern 83. Boulton AJ, Kirsner RS, Vileikyte L: Clinical 94. Papanas N, Papatheodorou Med 158:157–162, 1998 practice. Neuropathic diabetic foot ulcers. K, Papazoglou D, Christakidis 105. Frykberg RG, Lavery LA, Pham H, Harvey N Engl J Med 351:48–55, 2004 D, Monastiriotis C, Maltezos E: C, Harkless L, Veves A: Role of neurop- 84. Papanas N, Gries A, Maltezos E, Zick R: Reproducibility of the new indicator test athy and high foot pressures in diabetic The steel ball-bearing test: a new test for for sudomotor function (Neuropad) in foot ulceration. Diabetes Care 21:1714– evaluating protective sensation in the patients with type 2 diabetes mellitus: 1719, 1998 diabetic foot. Diabetologia 49:739–743, short communication. Exp Clin Endocrinol 106. Gonzalez R, Pedro T, Real JT, Martinez- 2006 Diabetes 113:577–581, 2005 Hervas S, Abellan MR, Lorente R, Priego A, 85. Carrington AL, Shaw JE, Van Schie CH, 95. Tentolouris N, Voulgari C, Liatis S, Catala M, Chaves FJ, Ascaso JF, Carmena Abbott CA, Vileikyte L, Boulton AJ: Can Kokkinos A, Eleftheriadou I, Makrilakis R: Plasma levels are associ- motor nerve conduction velocity predict K, Marinou K, Katsilambros N: Moisture ated with ulceration of the foot in patients foot problems in diabetic subjects over status of the skin of the feet assessed by with type 2 diabetes mellitus. Diabetes a 6-year outcome period? Diabetes Care the visual test neuropad correlates with Metab Res Rev 26:115–120, 2010 25:2010–2015, 2002 foot ulceration in diabetes. Diabetes Care 107. Jirkovska A, Boucek P, Woskova V, Bartos 86. Olmos PR, Cataland S, O’Dorisio TM, 33:1112–1114, 2010 V, Skibov J: Identification of patients at risk Casey CA, Smead WL, Simon SR: The 96. Boyko EJ, Ahroni JH, Cohen V, Nelson for diabetic foot: a comparison of stan- Semmes-Weinstein monofilament as KM, Heagerty PJ: Prediction of diabetic dardized noninvasive testing with routine a potential predictor of foot ulceration foot ulcer occurrence using commonly practice at community diabetes clinics. J in patients with noninsulin-dependent available clinical information: the Seattle Diabetes Complications 15:63–68, 2001 diabetes. Am J Med Sci 309:76–82, 1995 Diabetic Foot Study. Diabetes Care 108. Ledoux WR, Shofer JB, Smith DG, 87. Sauseng S, Kastenbauer T, Sokol G, Irsigler 29:1202–1207, 2006 Sullivan K, Hayes SG, Assal M, Reiber K: Estimation of risk for plantar foot ulcer- 97. Monteiro-Soares M, Dinis-Ribeiro M: GE: Relationship between foot type, foot ation in diabetic patients with neuropathy. External validation and optimisation deformity, and ulcer occurrence in the Diabetes Nutr Metab 12:189–193, 1999 of a model for predicting foot ulcers high-risk diabetic foot. J Rehabil Res Dev 88. Abbott CA, Carrington AL, Ashe H, Bath in patients with diabetes. Diabetologia 42:665–672, 2005 S, Every LC, Griffiths J, Hann AW, Hussein 53:1525–1533, 2010 109. Litzelman DK, Marriott DJ, Vinicor F: A, Jackson N, Johnson KE, Ryder CH, 98. Armstrong DG, Lavery LA: Elevated peak Independent physiological predictors Torkington R, Van Ross ER, Whalley plantar pressures in patients who have of foot lesions in patients with NIDDM. AM, Widdows P, Williamson S, Boulton Charcot arthropathy. J Bone Joint Surg Diabetes Care 20:1273–1278, 1997 AJ: The North-West Diabetes Foot Care Am 80:365–369, 1998 Study: incidence of, and risk factors for,

20–29 DIABETES IN AMERICA, 3rd Edition

110. Margolis DJ, Hofstad O, Feldman HI: 121. Meijer JW, Smit AJ, Sonderen EV, 133. Barbosa AP, Medina JL, Ramos EP, Barros Association between renal failure and Groothoff JW, Eisma WH, Links TP: HP: Prevalence and risk factors of clinical foot ulcer or lower-extremity amputation Symptom scoring systems to diagnose diabetic polyneuropathy in a Portuguese in patients with diabetes. Diabetes Care distal polyneuropathy in diabetes: the primary health care population. Diabetes 31:1331–1336, 2008 Diabetic Neuropathy Symptom score. Metab 27:496–502, 2001 111. Pham H, Armstrong DG, Harvey Diabet Med 19:962–965, 2002 134. Monami M, Vivarelli M, Desideri CM, C, Harkless LB, Giurini JM, Veves A: 122. Boulton AJ, Vileikyte L, Ragnarson- Colombi C, Marchionni N, Mannucci E: Screening techniques to identify people Tennvall G, Apelqvist J: The global and prediction of incident at high risk for diabetic foot ulceration: burden of diabetic foot disease. Lancet foot ulcers in type 2 diabetes. Diabetes a prospective multicenter trial. Diabetes 366:1719–1724, 2005 Care 32:897–899, 2009 Care 23:606–611, 2000 123. Lunetta M, Le Moli R, Grasso G, 135. Ndip A, Rutter MK, Vileikyte L, Vardhan 112. Saltzman CL, Rashid R, Hayes A, Fellner Sangiorgio L: A simplified diagnostic test A, Asari A, Jameel M, Tahir HA, Lavery C, Fitzpatrick D, Klapach A, Frantz R, Hillis for ambulatory screening of peripheral LA, Boulton AJ: Dialysis treatment is an SL: 4.5-gram monofilament sensation diabetic neuropathy. Diabetes Res Clin independent risk factor for foot ulceration beneath both first metatarsal heads Pract 39:165–172, 1998 in patients with diabetes and stage 4 or indicates protective foot sensation in 124. Moghtaderi A, Bakhshipour A, Rashidi 5 . Diabetes Care diabetic patients. J Bone Joint Surg Am H: Validation of Michigan neuropathy 33:1811–1816, 2010 86-A:717–723, 2004 screening instrument for diabetic periph- 136. Uccioli L, Faglia E, Monticone G, Favales 113. Suico JG, Marriott DJ, Vinicor F, Litzelman eral neuropathy. Clin Neurol Neurosurg F, Durola L, Aldeghi A, Quarantiello A, DK: Behaviors predicting foot lesions 108:477–481, 2006 Calia P, Menzinger G: Manufactured shoes in patients with non-insulin-dependent 125. Boulton AJ, Meneses P, Ennis WJ: in the prevention of diabetic foot ulcers. diabetes mellitus. J Gen Intern Med Diabetic foot ulcers: a framework for Diabetes Care 18:1376–1378, 1995 13:482–484, 1998 prevention and care. Wound Repair Regen 137. Guerrero-Romero F, Rodriguez-Moran M: 114. Tentolouris N, Marinou K, Kokotis 7:7–16, 1999 Relationship of microalbuminuria with P, Karanti A, Diakoumopoulou E, 126. Singh N, Armstrong DG, Lipsky BA: the diabetic foot ulcers in type II diabetes. Katsilambros N: Sudomotor dysfunction Preventing foot ulcers in patients with J Diabetes Complications 12:193–196, is associated with foot ulceration in diabetes. JAMA 293:217–228, 2005 1998 diabetes. Diabet Med 26:302–305, 2009 127. Faglia E, Favales F, Morabito A: New 138. Diouri A, Slaoui Z, Chadli A, El Ghomari 115. Bennett PJ, Stocks AE, Whittam DJ: ulceration, new major amputation, and H, Kebbou M, Marouan F, Farouqi A, Analysis of risk factors for neuropathic survival rates in diabetic subjects hospi- Ababou MR: [Incidence of factors favoring foot ulceration in diabetes mellitus. J Am talized for foot ulceration from 1990 to recurrent foot ulcers in diabetic patients]. Podiatr Med Assoc 86:112–116, 1996 1993: a 6.5-year follow-up. Diabetes Care [Article in French] Ann Endocrinol (Paris) 116. Caselli A, Pham H, Giurini JM, Armstrong 24:78–83, 2001 63:491–496, 2002 DG, Veves A: The forefoot-to-rearfoot 128. Centers for Disease Control and 139. Armstrong DG, Lavery LA, Liswood PJ, plantar pressure ratio is increased in Prevention (CDC): History of foot ulcer Todd WF, Tredwell JA: Infrared dermal severe diabetic neuropathy and can among persons with diabetes—United thermometry for the high-risk diabetic predict foot ulceration. Diabetes Care States, 2000–2002. MMWR Morb Mortal foot. Phys Ther 77:169–175; discussion 25:1066–1071, 2002 Wkly Rep 52:1098–1102, 2003 176–177, 1997 117. Gonzalez JS, Vileikyte L, Ulbrecht 129. Murray HJ, Young MJ, Hollis S, Boulton 140. Abbott CA, Garrow AP, Carrington AL, JS, Rubin RR, Garrow AP, Delgado C, AJ: The association between callus Morris J, Van Ross ER, Boulton AJ; North- Cavanagh PR, Boulton AJ, Peyrot M: formation, high pressures and neuropathy West diabetes foot care study: Foot ulcer Depression predicts first but not recur- in diabetic foot ulceration. Diabet Med risk is lower in South-Asian and African- rent diabetic foot ulcers. Diabetologia 13:979–982, 1996 Caribbean compared with European 53:2241–2248, 2010 130. Armstrong DG, Peters EJ, Athanasiou diabetic patients in the U.K.: the North- 118. Maluf KS, Mueller MJ: Novel Award KA, Lavery LA: Is there a critical level of West diabetes foot care study. Diabetes 2002. Comparison of physical activity plantar foot pressure to identify patients Care 28:1869–1875, 2005 and cumulative plantar tissue stress at risk for neuropathic foot ulceration? J 141. Ndip A, Lavery LA, Lafontaine J, Rutter among subjects with and without diabetes Foot Ankle Surg 37:303–307, 1998 MK, Vardhan A, Vileikyte L, Boulton AJ: mellitus and a history of recurrent plantar 131. Lavery LA, Armstrong DG, Wunderlich High levels of foot ulceration and amputa- ulcers. Clin Biomech (Bristol, Avon) RP, Tredwell J, Boulton AJ: Predictive tion risk in a multiracial cohort of diabetic 18:567–575, 2003 value of foot pressure assessment as part patients on dialysis therapy. Diabetes 119. Miranda-Palma B, Sosenko JM, Bowker of a population-based diabetes disease Care 33:878–880, 2010 JH, Mizel MS, Boulton AJ: A comparison management program. Diabetes Care 142. Jeffcoate W: New guidelines for the manage- of the monofilament with other testing 26:1069–1073, 2003 ment of the diabetic foot in hospitals: so modalities for foot ulcer susceptibility. 132. Armstrong DG, Lavery LA, Holtz-Neiderer far so good... but will we get Cinderella to Diabetes Res Clin Pract 70:8–12, 2005 K, Mohler MJ, Wendel CS, Nixon BP, the ball? Diabet Med 29:2–4, 2012 120. Boulton AJ, Kubrusly DB, Bowker JH, Boulton AJ: Variability in activity may 143. Armstrong DG, Lavery LA, Harkless LB: Gadia MT, Quintero L, Becker DM, Skyler precede diabetic foot ulceration. Diabetes Treatment-based classification system for JS, Sosenko JM: Impaired vibratory Care 27:1980–1984, 2004 assessment and care of diabetic feet. J perception and diabetic foot ulceration. Am Podiatr Med Assoc 86:311–316, 1996 Diabet Med 3:335–337, 1986 144. Monteiro-Soares M, Martins-Mendes D, Dinis-Ribeiro M, Guimaraes R, Tavora A, Lemos E, Sobral J, Duarte I,

20–30 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

Campos-Lemos J, Brandao D, Madureira Conventional wisdom or evidence-based in individuals with diabetic foot ulcers: M, Ribeiro M, Oliveira M: The impact practice? Diabetes Care 20:156–162, focus on the differences between of impaired renal function on diabetic 1997 individuals with and without peripheral foot complications prediction: should it 156. Boulton AJ: The diabetic foot: a global arterial disease. The EURODIALE Study. be included on foot risk classifications? view. Diabetes Metab Res Rev 16(Suppl Diabetologia 51:747–755, 2008 Diabetologia 56(Suppl1):S501, 2013 1):S2–S5, 2000 167. Armstrong DG, Lavery LA, Harkless LB: 145. Monteiro-Soares M, Boyko EJ, Ribeiro 157. Chantelau E, Kushner T, Spraul M: How Validation of a diabetic wound classifica- J, Ribeiro I, Dinis-Ribeiro M: Risk effective is cushioned therapeutic foot- tion system. The contribution of depth, stratification systems for diabetic foot wear in protecting diabetic feet? A clinical infection, and ischemia to risk of amputa- ulcers: a systematic review. Diabetologia study. Diabet Med 7:355–359, 1990 tion. Diabetes Care 21:855–859, 1998 54:1190–1199, 2011 158. Reiber GE, Smith DG, Wallace C, Sullivan 168. Beckert S, Witte M, Wicke C, Konigsrainer 146. Armstrong DG, Holtz K, Wu S: Can the K, Hayes S, Vath C, Maciejewski ML, Yu A, Coerper S: A new wound-based use of a topical antifungal nail lacquer O, Heagerty PJ, LeMaster J: Effect of severity score for diabetic foot ulcers: a reduce risk for diabetic foot ulceration? therapeutic footwear on foot reulceration prospective analysis of 1,000 patients. Results from a randomised controlled in patients with diabetes: a randomized Diabetes Care 29:988–992, 2006 pilot study. Int Wound J 2:166–170, 2005 controlled trial. JAMA 287:2552–2558, 169. Malay DS, Margolis DJ, Hoffstad OJ, 147. Martins-Mendes D, Monteiro-Soares M, 2002 Bellamy S: The incidence and risks of Boyko EJ, Ribeiro M, Barata P, Lima J, 159. Chantelau E, Haage P: An audit of cush- failure to heal after lower extremity Soares R: The independent contribution ioned diabetic footwear: relation to patient amputation for the treatment of diabetic of diabetic foot ulcer on lower extremity compliance. Diabet Med 11:114–116, 1994 neuropathic foot ulcer. J Foot Ankle Surg amputation and mortality risk. J Diabetes 160. Armstrong DG, Lavery LA, Wunderlich RP, 45:366–374, 2006 Complications 28:632–638, 2014 Boulton AJ: 2003 William J. Stickel Silver 170. Martinez-De Jesus FR: A checklist system 148. Mayfield JA, Reiber GE, Sanders LJ, Award. Skin temperatures as a one-time to score healing progress of diabetic foot Janisse D, Pogach LM: Preventive foot screening tool do not predict future ulcers. Int J Low Extrem Wounds 9:74–83, care in diabetes. Diabetes Care 27(Suppl diabetic foot complications. J Am Podiatr 2010 1):S63–S64, 2004 Med Assoc 93:443–447, 2003 171. Abbas ZG, Lutale JK, Game FL, Jeffcoate 149. Plank J, Haas W, Rakovac I, Gorzer E, 161. Armstrong DG, Holtz-Neiderer K, Wendel WJ: Comparison of four systems of classi- Sommer R, Siebenhofer A, Pieber TR: C, Mohler MJ, Kimbriel HR, Lavery LA: fication of diabetic foot ulcers in Tanzania. Evaluation of the impact of chiropodist Skin temperature monitoring reduces the Diabet Med 25:134–137, 2008 care in the secondary prevention of foot risk for diabetic foot ulceration in high- 172. Leese G, Schofield C, McMurray B, Libby ulcerations in diabetic subjects. Diabetes risk patients. Am J Med 120:1042–1046, G, Golden J, MacAlpine R, Cunningham S, Care 26:1691–1695, 2003 2007 Morris A, Flett M, Griffiths G: Scottish foot 150. Calle-Pascual AL, Duran A, Benedi A, 162. Lavery LA, Higgins KR, Lanctot DR, ulcer risk score predicts foot ulcer healing Calvo MI, Charro A, Diaz JA, Calle JR, Gil Constantinides GP, Zamorano RG, in a regional specialist foot clinic. Diabetes E, Ibarra J, Maranes JP, Cabezas-Cerrato J: Armstrong DG, Athanasiou KA, Agrawal Care 30:2064–2069, 2007 Reduction in foot ulcer incidence: relation CM: Home monitoring of foot skin 173. Margolis DJ, Allen-Taylor L, Hoffstad to compliance with a prophylactic foot temperatures to prevent ulceration. O, Berlin JA: Diabetic neuropathic foot care program. Diabetes Care 24:405–407, Diabetes Care 27:2642–2647, 2004 ulcers: predicting which ones will not heal. 2001 163. Lavery LA, Higgins KR, Lanctot DR, Am J Med 115:627–631, 2003 151. Lincoln NB, Radford KA, Game FL, Constantinides GP, Zamorano RG, 174. Parisi MC, Zantut-Wittmann DE, Pavin Jeffcoate WJ: Education for secondary Athanasiou KA, Armstrong DG, Agrawal EJ, Machado H, Nery M, Jeffcoate WJ: prevention of foot ulcers in people with CM: Preventing diabetic foot ulcer Comparison of three systems of classifica- diabetes: a randomised controlled trial. recurrence in high-risk patients: use of tion in predicting the outcome of diabetic Diabetologia 51:1954–1961, 2008 temperature monitoring as a self-assess- foot ulcers in a Brazilian population. Eur J 152. Dorresteijn JA, Kriegsman DM, ment tool. Diabetes Care 30:14–20, 2007 Endocrinol 159:417–422, 2008 Assendelft WJ, Valk GD: Patient 164. Houghton VJ, Bower VM, Chant DC: Is an 175. Treece KA, Macfarlane RM, Pound N, education for preventing diabetic foot increase in skin temperature predictive Game FL, Jeffcoate WJ: Validation of ulceration. Cochrane Database Syst Rev of neuropathic foot ulceration in people a system of foot ulcer classification in 12:CD001488, 2014 with diabetes? A systematic review and diabetes mellitus. Diabet Med 21:987– 153. Monteiro-Soares M, Vaz-Carneiro A, meta-analysis. J Foot Ankle Res 6:31, 991, 2004 Sampaio S, Dinis-Ribeiro M: Validation 2013 176. Ince P, Game FL, Jeffcoate WJ: Rate of and comparison of currently available 165. Basu S, Hadley J, Tan RM, Williams J, healing of neuropathic ulcers of the foot stratification systems for patients with Shearman CP: Is there enough infor- in diabetes and its relationship to ulcer diabetes by risk of foot ulcer development. mation about foot care among patients duration and ulcer area. Diabetes Care Eur J Endocrinol 167:401–407, 2012 with diabetes? Int J Low Extrem Wounds 30:660–663, 2007 154. Busch K, Chantelau E: Effectiveness of 3:64–68, 2004 177. Margolis DJ, Allen-Taylor L, Hoffstad O, a new brand of stock “diabetic” shoes to 166. Prompers L, Schaper N, Apelqvist J, Berlin JA: Diabetic neuropathic foot ulcers: protect against diabetic foot ulcer relapse. Edmonds M, Jude E, Mauricio D, Uccioli the association of wound size, wound A prospective cohort study. Diabet Med L, Urbancic V, Bakker K, Holstein P, duration, and wound grade on healing. 20:665–669, 2003 Jirkovska A, Piaggesi A, Ragnarson- Diabetes Care 25:1835–1839, 2002 155. Litzelman DK, Marriott DJ, Vinicor F: Tennvall G, Reike H, Spraul M, Van Acker 178. Oyibo SO, Jude EB, Tarawneh I, Nguyen The role of footwear in the prevention K, Van Baal J, Van Merode F, Ferreira HC, Armstrong DG, Harkless LB, Boulton of foot lesions in patients with NIDDM. I, Huijberts M: Prediction of outcome AJ: The effects of ulcer size and site,

20–31 DIABETES IN AMERICA, 3rd Edition

patient’s age, sex and type and duration of debriding and off-loading the diabetic foot 202. Johannesson A, Larsson GU, Ramstrand diabetes on the outcome of diabetic foot wound. Clin Infect Dis 39(Suppl 2):S92– N, Turkiewicz A, Wirhn AB, Atroshi I: ulcers. Diabet Med 18:133–138, 2001 S99, 2004 Incidence of lower-limb amputation in the 179. Lavery LA, Barnes SA, Keith MS, Seaman 190. Lebrun E, Tomic-Canic M, Kirsner RS: The diabetic and nondiabetic general popula- JW, Jr., Armstrong DG: Prediction of role of surgical debridement in healing of tion: a 10-year population-based cohort healing for postoperative diabetic foot diabetic foot ulcers. Wound Repair Regen study of initial unilateral and contralateral wounds based on early wound area 18:433–438, 2010 amputations and reamputations. Diabetes progression. Diabetes Care 31:26–29, 191. Armstrong DG, Boulton AJ: Pressure Care 32:275–280, 2009 2008 offloading and “advanced” wound 203. Young BA, Maynard C, Reiber G, Boyko 180. Adler AI, Boyko EJ, Ahroni JH, Smith DG: healing: isn’t it finally time for an arranged EJ: Effects of ethnicity and nephropathy Lower-extremity amputation in diabetes. marriage? Int J Low Extrem Wounds on lower-extremity amputation risk The independent effects of peripheral 3:184–187, 2004 among diabetic veterans. Diabetes Care vascular disease, sensory neuropathy, 192. Cavanagh PR, Bus SA: Off-loading the 26:495–501, 2003 and foot ulcers. Diabetes Care 22:1029– diabetic foot for ulcer prevention and 204. Li Y, Burrows NR, Gregg EW, Albright 1035, 1999 healing. Plast Reconstr Surg 127(Suppl A, Geiss LS: Declining rates of hospital- 181. Edelman D, Hough DM, Glazebrook KN, 1):248S–256S, 2011 ization for nontraumatic lower-extremity Oddone EZ: Prognostic value of the clin- 193. Lewis J, Lipp A: Pressure-relieving amputation in the diabetic population ical examination of the diabetic foot ulcer. interventions for treating diabetic foot aged 40 years or older: U.S., 1988–2008. J Gen Intern Med 12:537–543, 1997 ulcers. Cochrane Database Syst Rev Diabetes Care 35:273–277, 2012 182. Lepantalo M, Apelqvist J, Setacci C, Ricco 1:CD002302, 2013 205. Data Points Publication Series [Internet]: JB, de Donato G, Becker F, Robert-Ebadi 194. Dang CN, Boulton AJ: Changing perspec- Prevalence of diabetes, diabetic foot ulcer, H, Cao P, Eckstein HH, De Rango P, tives in diabetic foot ulcer management. and lower extremity amputation among Diehm N, Schmidli J, Teraa M, Moll FL, Int J Low Extrem Wounds 2:4–12, 2003 Medicare beneficiaries, 2006 to 2008: Dick F, Davies AH: Chapter V: Diabetic 195. Rathur HM, Boulton AJ: The diabetic foot. Data Points #1 [online article], 2011. foot. Eur J Vasc Endovasc Surg 42(Suppl Clin Dermatol 25:109–120, 2007 Available from http://www.ncbi.nlm.nih. 2):S60–S74, 2011 196. Tsourdi E, Barthel A, Rietzsch H, Reichel gov/books/NBK63602. 183. Monteiro-Soares M, Martins-Mendes A, Bornstein SR: Current aspects in the 206. Wrobel JS, Mayfield JA, Reiber GE: D, Vaz-Carneiro A, Sampaio S, Dinis- pathophysiology and treatment of chronic Geographic variation of lower-extremity Ribeiro M: Classification systems for wounds in diabetes mellitus. Biomed Res major amputation in individuals with and lower extremity amputation prediction in Int 2013:385641, 2013 without diabetes in the Medicare popula- subjects with active diabetic foot ulcer: 197. Abidia A, Laden G, Kuhan G, Johnson tion. Diabetes Care 24:860–864, 2001 a systematic review and meta-analysis. BF, Wilkinson AR, Renwick PM, Masson 207. Holman N, Young RJ, Jeffcoate WJ: Diabetes Metab Res Rev 30:610–622, EA, McCollum PT: The role of hyperbaric Variation in the recorded incidence of 2014 oxygen therapy in ischaemic diabetic amputation of the lower limb in England. 184. Karthikesalingam A, Holt PJ, Moxey P, lower extremity ulcers: a double-blind Diabetologia 55:1919–1925, 2012 Jones KG, Thompson MM, Hinchliffe RJ: A randomised-controlled trial. Eur J Vasc 208. Goldberg JB, Goodney PP, Cronenwett systematic review of scoring systems for Envasc Surg 25:513–518, 2003 JL, Baker F: The effect of risk and race diabetic foot ulcers. Diabet Med 27:544– 198. Liu R, Li L, Yang M, Boden G, Yang G: on lower extremity amputations among 549, 2010 Systematic review of the effectiveness Medicare diabetic patients. J Vasc Surg 185. Leese GP, Reid F, Green V, McAlpine of hyperbaric oxygenation therapy in the 56:1663–1668, 2012 R, Cunningham S, Emslie-Smith AM, management of chronic diabetic foot 209. Wallace GF: Indications for amputations. Morris AD, McMurray B, Connacher AC: ulcers. Mayo Clin Proc 88:166–175, 2013 Clin Podiatr Med Surg 22:315–328, 2005 Stratification of foot ulcer risk in patients 199. Londahl M, Katzman P, Nilsson A, 210. Pecoraro RE, Reiber GE, Burgess EM: with diabetes: a population-based study. Hammarlund C: Hyperbaric oxygen Pathways to diabetic limb amputation. Int J Clin Pract 60:541–545, 2006 therapy facilitates healing of chronic foot Basis for prevention. Diabetes Care 186. Armstrong DG, Lavery LA: Decreasing ulcers in patients with diabetes. Diabetes 13:513–521, 1990 foot pressures while implementing Care 33:998–1003, 2010 211. Adler AI, Erqou S, Lima TA, Robinson AH: topical negative pressure (vacuum-as- 200. Centers for Disease Control and Association between glycated haemo- sisted closure) therapy. Int J Low Extrem Prevention (CDC): Hospital discharge globin and the risk of lower extremity Wounds 3:12–15, 2004 rates for nontraumatic lower extremity amputation in patients with diabetes 187. Jeffcoate WJ, Rasmussen LM, Hofbauer amputation by diabetes status—United mellitus—review and meta-analysis. LC, Game FL: Medial arterial calcification States, 1997. MMWR Morb Mort Wkly Diabetologia 53:840–849, 2010 in diabetes and its relationship to neurop- Rep 50:954–958, 2001 212. Selby JV, Zhang D: Risk factors for lower athy. Diabetologia 52:2478–2488, 2009 201. Resnick HE, Valsania P, Phillips CL: extremity amputation in persons with 188. Armstrong DG, Lavery LA: Negative Diabetes mellitus and nontraumatic diabetes. Diabetes Care 18:509–516, pressure wound therapy after partial lower extremity amputation in black and 1995 diabetic foot amputation: a multicentre, white Americans: the National Health 213. Chaturvedi N, Abbott CA, Whalley A, randomised controlled trial. Lancet and Nutrition Examination Survey Widdows P, Leggetter SY, Boulton AJ: Risk 366:1704–1710, 2005 Epidemiologic Follow-up Study, 1971– of diabetes-related amputation in South 189. Armstrong DG, Lavery LA, Nixon BP, 1992. Arch Intern Med 159:2470–2475, Asians vs. Europeans in the UK. Diabet Boulton AJ: It’s not what you put on, 1999 Med 19:99–104, 2002 but what you take off: techniques for

20–32 Peripheral Arterial Disease, Foot Ulcers, Lower Extremity Amputations, and Diabetes

214. Chaturvedi N, Stevens LK, Fuller JH, Lee 225. Davis WA, Norman PE, Bruce DG, Davis ET, Lu M: Risk factors, ethnic differ- TM: Predictors, consequences and costs ences and mortality associated with of diabetes-related lower extremity ampu- lower-extremity gangrene and amputation tation complicating type 2 diabetes: the in diabetes. The WHO Multinational Fremantle Diabetes Study. Diabetologia Study of Vascular Disease in Diabetes. 49:2634–2641, 2006 Diabetologia 44(Suppl 2):S65–S71, 2001 215. Leggetter S, Chaturvedi N, Fuller JH, Edmonds ME: Ethnicity and risk of diabe- tes-related lower extremity amputation: a population-based, case-control study of African Caribbeans and Europeans in the United Kingdom. Arch Intern Med 162:73–78, 2002 216. Callaghan BC, Feldman E, Liu J, Kerber K, Pop-Busui R, Moffet H, Karter AJ: and amputation risk in patients with diabetes: ten-year follow-up in the DISTANCE study. Diabetes Care 34:635–640, 2011 217. Agarwal S: The association of active and passive smoking with peripheral arterial disease: results from NHANES 1999– 2004. 60:335–345, 2009 218. Borkosky SL, Roukis TS: Incidence of re-amputation following partial first ray amputation associated with diabetes mellitus and peripheral sensory neurop- athy: a systematic review. Diabet Foot Ankle 3, 2012 219. Izumi Y, Satterfield K, Lee S, Harkless LB: Risk of reamputation in diabetic patients stratified by limb and level of amputation: a 10-year observation. Diabetes Care 29:566–570, 2006 220. Skoutas D, Papanas N, Georgiadis GS, Zervas V, Manes C, Maltezos E, Lazarides MK: Risk factors for ipsilateral reamputa- tion in patients with diabetic foot lesions. Int J Low Extrem Wounds 8:69–74, 2009 221. Ahroni JH, Boyko EJ, Davignon DR, Pecoraro RE: The health and functional status of veterans with diabetes. Diabetes Care 17:318–321, 1994 222. Al Snih S, Fisher MN, Raji MA, Markides KS, Ostir GV, Goodwin JS: Diabetes mellitus and incidence of lower body disability among older Mexican Americans. J Gerontol A Biol Sci Med Sci 60:1152–1156, 2005 223. Peters EJ, Childs MR, Wunderlich RP, Harkless LB, Armstrong DG, Lavery LA: Functional status of persons with diabe- tes-related lower-extremity amputations. Diabetes Care 24:1799–1804, 2001 224. Tentolouris N, Al-Sabbagh S, Walker MG, Boulton AJ, Jude EB: Mortality in diabetic and nondiabetic patients after amputa- tions performed from 1990 to 1995: a 5-year follow-up study. Diabetes Care 27:1598–1604, 2004

20–33 DIABETES IN AMERICA, 3rd Edition

APPENDICES

APPENDIX 20.1. Mean Body Mass Index (kg/m2) Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

MEAN (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 30.0 (0.5) 31.4 (0.3) 30.5 (0.5) 31.3 (0.4) 28.9 (0.8) 31.4 (0.5) 27.9 (0.8) 27.9 (0.2) Age (years) 40–64 31.0 (1.2) 32.2 (0.4) 31.9 (0.9) 32.1 (0.5) 29.4 (1.7) 32.5 (0.7) 30.3 (1.7) 28.0 (0.2) 65–74 30.9 (0.9) 30.5 (0.4) 31.5 (1.1) 30.7 (0.4) 29.3 (1.0) 29.9 (0.7) 26.1 (0.8) 28.1 (0.3) ≥75 27.5 (0.5) 28.4 (0.4) 27.2 (0.7) 28.5 (0.5) 28.0 (0.6) 28.0 (0.9) 26.4 (0.9) 26.1 (0.3) Sex Men 29.4 (0.5) 30.7 (0.4) 30.0 (0.6) 30.8 (0.5) 28.3 (0.7) 30.5 (0.5) 26.2 (0.6) 28.1 (0.2) Women 30.9 (0.7) 32.2 (0.5) 31.2 (0.8) 31.9 (0.5) 30.1 (1.2) 32.8 (1.2) 29.1 (1.3) 27.7 (0.3) Race/ethnicity Non-Hispanic white 30.2 (0.6) 31.6 (0.5) 31.1 (0.7) 31.9 (0.5) 28.7 (0.8) 31.1 (0.6) 27.9 (1.0) 27.8 (0.2) Non-Hispanic black 30.3 (0.8) 32.2 (0.4) 30.5 (0.9) 31.7 (0.5) 29.4 (1.7) 33.5 (0.8) 29.3 (1.1) 29.4 (0.3) All Hispanic 27.9 (1.9) 30.3 (0.4) 26.5 (2.2) 29.8 (0.5) 30.3 (2.2) 31.8 (0.8) 28.3 (0.9) 28.2 (0.3) Mexican American 31.0 (0.9) 30.9 (0.3) 29.9 (0.9) 30.3 (0.4) 34.6 (1.6) 32.6 (0.6) 28.3 (1.4) 28.7 (0.2) Body mass index is based on measured height and weight. Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. All relative standard errors ≤30% SOURCE: National Health and Nutrition Examination Surveys 1999–2004

APPENDIX 20.2. Mean Waist Circumference (cm) Among Adults Age ≥40 Years, Overall and by Age, Sex, Race/Ethnicity, Peripheral Arterial Disease, and Diabetes Status, U.S., 1999–2004

MEAN (STANDARD ERROR) All Diabetes Diagnosed Diabetes Undiagnosed Diabetes No Diabetes CHARACTERISTICS PAD No PAD PAD No PAD PAD No PAD PAD No PAD Overall 106.7 (1.2) 107.4 (0.8) 107.9 (1.5) 107.3 (0.9) 104.4 (2.0) 107.6 (1.1) 98.6 (1.4) 97.0 (0.5) Age (years) 40–64 109.0 (3.0) 108.3 (1.1) 112.6 (2.2) 108.1 (1.3) 103.3 (4.0) 108.8 (1.5) 101.2 (3.0) 96.8 (0.6) 65–74 108.4 (2.1) 107.4 (0.9) 109.5 (2.7) 107.5 (1.1) 105.9 (2.6) 107.1 (1.7) 96.8 (2.2) 99.1 (0.9) ≥75 101.9 (1.4) 102.2 (0.9) 100.9 (1.9) 102.3 (1.0) 103.9 (1.3) 102.0 (2.1) 96.4 (1.7) 95.6 (0.8) Sex Men 108.2 (1.8) 108.5 (1.1) 110.6 (1.9) 108.7 (1.4) 104.2 (2.5) 108.0 (1.5) 99.8 (1.4) 101.8 (0.5) Women 104.7 (1.5) 106.0 (1.1) 104.7 (2.0) 105.7 (1.2) 104.8 (2.3) 106.9 (2.4) 97.8 (2.4) 92.7 (0.7) Race/ethnicity Non-Hispanic white 107.6 (1.5) 109.1 (1.2) 109.9 (1.8) 109.6 (1.3) 103.9 (2.3) 108.2 (1.4) 98.9 (1.7) 97.4 (0.6) Non-Hispanic black 106.9 (1.4) 106.7 (0.9) 106.7 (1.7) 106.2 (1.1) 107.7 (2.7) 108.2 (1.6) 98.8 (2.6) 97.3 (0.7) All Hispanic 99.1 (4.6) 102.8 (1.0) 95.3 (5.4) 101.7 (1.1) 105.7 (3.3) 105.8 (2.3) 99.4 (2.4) 95.9 (0.8) Mexican American 105.7 (2.0) 104.5 (0.7) 103.9 (2.8) 103.4 (1.0) 111.5 (3.8) 107.0 (1.5) 98.7 (3.1) 97.2 (0.6) Peripheral arterial disease (PAD) is defined as ankle-brachial index <0.9 on either leg. Diagnosed diabetes is based on self-report. Undiagnosed diabetes is defined as A1c ≥6.5% or fasting plasma glucose ≥126 mg/dL. Conversions for A1c and glucose values are provided in Diabetes in America Appendix 1 Conversions. A1c, glycosylated hemoglobin. All relative standard errors ≤30% SOURCE: National Health and Nutrition Examination Surveys 1999–2004

20–34